Type:	Package
Title:	Exploration and Graphics for RivEr Trends
Version:	3.0.11
Description:	Statistics and graphics for streamflow history, water quality trends, and the statistical modeling algorithm: Weighted Regressions on Time, Discharge, and Season (WRTDS).
License:	CC0
Depends:	R (≥ 3.5)
Imports:	dataRetrieval (≥ 2.0.1), survival, fields, methods, utils, grDevices, graphics, stats, truncnorm, foreach, MASS
Suggests:	EGRETci, knitr, rmarkdown, extrafont, testthat, rkt, doParallel, parallel, pkgdown, png, dplyr, zyp, lubridate, covr
LazyLoad:	yes
LazyData:	yes
BugReports:	https://github.com/DOI-USGS/EGRET/issues
VignetteBuilder:	knitr
BuildVignettes:	true
URL:	https://pubs.usgs.gov/tm/04/a10/
Copyright:	This software is in the public domain because it contains materials that originally came from the United States Geological Survey, an agency of the United States Department of Interior. For more information, see the official USGS copyright policy at https://www.usgs.gov/information-policies-and-instructions/copyrights-and-credits
RoxygenNote:	7.3.2
Encoding:	UTF-8
NeedsCompilation:	no
Packaged:	2025-04-29 21:08:54 UTC; ldecicco
Author:	Robert Hirsch [aut], Laura DeCicco [aut, cre], Tim Cohn [ctb], David Watkins [ctb], Lindsay Carr [ctb], Jennifer Murphy [aut]
Maintainer:	Laura DeCicco <ldecicco@usgs.gov>
Repository:	CRAN
Date/Publication:	2025-04-29 21:30:02 UTC

EGRET package includes WRTDS and flowHistory

Description

Details

Collection of functions to do WRTDS and flowHistory analysis, and produce graphs and tables of data and results from these analyses.

Author(s)

Robert M. Hirsch rhirsch@usgs.gov, Laura De Cicco ldecicco@usgs.gov

References

Hirsch, R.M., and De Cicco, L.A., 2014, User guide to Exploration and Graphics for RivEr Trends (EGRET) and dataRetrieval: R packages for hydrologic data: U.S. Geological Survey Techniques and Methods book 4, chap. A10, 94 p., doi:10.3133/tm4A10

See Also

Useful links:

• https://pubs.usgs.gov/tm/04/a10/

• Report bugs at https://github.com/DOI-USGS/EGRET/issues

Example eList

Description

Example data representing data from the Choptank River at Greensboro, MD, USGS data Data is a named list of the Daily, Sample, INFO dataframes, and the surface matrix.

Examples

head(Choptank_eList$Daily)
head(Arkansas_eList$Daily)

Constants included with EGRET

Description

Examples

fluxConst
fluxConst[['kgDay']]
fluxConst[['kgDay']]@unitName
qConst
qConst[['cfs']]
qConst[['cfs']]@qUnitName
concConst[['concentration']]
concConst[['concentration']]@shortPrefix

Import metadata to create INFO data frame

Description

Populates INFO data frame from either NWIS (readNWISInfo), Water Quality Portal (readWQPInfo), or user-supplied files (readUserInfo).

Usage

readNWISInfo(siteNumber, parameterCd, interactive = TRUE)

readWQPInfo(siteNumber, parameterCd, interactive = TRUE)

readUserInfo(filePath, fileName, hasHeader = TRUE, separator = ",",
  interactive = TRUE)

Arguments

Value

INFO data frame. Any metadata can be stored in INFO. However, there are 8 columns that EGRET uses by name in some functions:

*** Additionally, EGRET assumes that all concentrations are saved in mg/l. If some variation of 'mg/l' is not found in INFO$param.units, functions that calculate flux will issue a warning. This is because the conversion from mg/l to the user-specified flux unit (e.g., kg/day) uses hard-coded conversion factors.

See Also

readNWISsite, readNWISpCode

whatWQPsites

Examples

# These examples require an internet connection to run
# Automatically gets information about site 05114000 and temperature

INFO <- readNWISInfo('05114000','00010',interactive = FALSE)



# These examples require an internet connection to run
# Automatically gets information about site 01594440 and temperature, no interaction with user
nameToUse <- 'Specific conductance'
pcodeToUse <- '00095'

INFO <- readWQPInfo('USGS-04024315',pcodeToUse, interactive = FALSE)

INFO2 <- readWQPInfo('WIDNR_WQX-10032762',nameToUse, interactive = FALSE)
# To adjust the label names:
INFO$shortName <- "Little"
INFO$paramShortName <- "SC"


filePath <- system.file("extdata", package="EGRET")
fileName <- 'infoTest.csv'
INFO <- readUserInfo(filePath,fileName, separator=",",interactive=FALSE)

WRTDS-Kalman

Description

This function uses an autoregressive model to produce more accurate estimates of concentration and flux

Usage

WRTDSKalman(eList, rho = 0.9, niter = 200, seed = NA, verbose = TRUE)

Arguments

Details

This function takes an existing eList Including the estimated model (the surfaces object in the eList) And produces the daily WRTDSKalman estimates of concentration and flux These generated estimates are called genConc and genFlux

Examples

eList <- Choptank_eList
eList <- WRTDSKalman(eList, niter = 10)
summary(eList$Daily)

#All flux values in AnnualResults are expressed as a rate in kg/day
AnnualResults <- setupYears(eList$Daily)
head(AnnualResults)

Create named list for EGRET analysis

Description

Create a named list with the INFO, Daily, and Sample dataframes, and surface matrix. If any of these are not available, an NA should be

Usage

as.egret(INFO, Daily, Sample = NA, surfaces = NA)

Arguments

Value

eList named list with Daily, Sample, and INFO dataframes, along with the surfaces matrix. Any of these values can be NA, not all EGRET functions will work with missing parts of the named list eList.

See Also

readNWISDaily, readNWISSample

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)
INFO <- getInfo(eList)
eList_flowHistory <- as.egret(INFO, Daily)
plotFlowSingle(eList_flowHistory, 1)
Sample <- getSample(eList)
surfaces <- getSurfaces(eList)
eList_full <- as.egret(INFO, Daily, Sample, surfaces)
plotFluxQ(eList_full)

Deletes the computed values during periods of time when there are no sample data

Description

This function is used when the data analyst believes that a gap in the sample data record is so long that estimates during that period are not reliable. This is only used for periods of several years in duration. For this period, the values of Conc, Flux, FNConc and FNFlux are all converted to NA.

Usage

blankTime(eList, startBlank, endBlank)

Arguments

Details

The startBlank and endBlank arguments should generally coincide with the starting and ending date of the period of analysis that is being used. startBlank should be placed fairly close to the start of the period of no data and endBlank should be placed fairly close to the end of the period of no data. They do not eliminate any water quality data from the set of data being used to estimate the model, they only eliminate results computed for the specified blank period. If the data set has more than one large data gap the blankTime() function can be used multiple times to blank out multiple sets of results.

Value

eList named list with modified Daily data frame.

Examples

startBlank = "2004-10-01"
endBlank = "2006-09-30"
eList <- Choptank_eList
eList <- blankTime(eList, startBlank, endBlank)

Box plot of the water quality data by month

Description

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Box widths are proportional to the square root of the number of samples in the month.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

boxConcMonth(eList, printTitle = TRUE, cex = 0.8, cex.axis = 1.1,
  cex.main = 1.1, las = 1, logScale = FALSE, tcl = 0.5,
  tinyPlot = FALSE, customPar = FALSE, showYLabels = TRUE, concLab = 1,
  showXLabels = TRUE, showXAxis = TRUE, showYAxis = TRUE, monthLab = 1,
  ...)

Arguments

See Also

boxplot

Examples

eList <- Choptank_eList
# Water year:
boxConcMonth(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
boxConcMonth(eList)
spanish_month <- new("monthLabel",
                  monthAbbrev = c("enero",	"feb", 	"marzo", "abr",
                                  "mayo",	"jun",	"jul", "agosto", "set",
                                  "oct",	"nov", "dic"),
                  monthFull = c("enero",	"febrero", 	"marzo", "abril",
                                  "mayo",	"junio",	"julio", "agosto", "septiembre",
                                  "octubre",	"noviembre", "diciembre"),
                  monthSingle = c("E", "F", "M", "A", "M", "J", "J",
                                  "A", "S", "O", "N", "D"))
boxConcMonth(eList, monthLab = spanish_month, 
             showXLabels = FALSE, printTitle = FALSE)

Three box plots side-by-side

Description

This function is used to compare the distribution of concentration in the sample and predicted data set. It shows three boxplots. One for the sample, one for the predictions on days with sample values, and one for all days (whether or not they had sample values). Box widths are proportional to the square root of the number of observations represented by the box.

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

boxConcThree(eList, tinyPlot = FALSE, printTitle = TRUE,
  moreTitle = "WRTDS", customPar = FALSE, font.main = 2, cex = 0.8,
  cex.main = 1.1, cex.axis = 1.1, concLab = 1, ...)

Arguments

See Also

boxplot

Examples

eList <- Choptank_eList
# Water year:
boxConcThree(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
boxConcThree(eList)

Two box plots side-by-side, discharge on sample days, and discharge on all days

Description

This function is used to compare the distribution of discharges in the sample data set and the discharges in the full daily data set. Note that discharge is plotted on a logarithmic axis. The boxplot is created using the log values but the scale is presented in the original units. An ideal situation would show the two boxes roughly similar to each other or the sample boxplot having median, upper quartile, and higher values being slightly greater than in the boxplot of all days.

Box widths are proportional to the square root of the number of observations (left box based on number of sampled days, right box based on total number of days in the record).

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

boxQTwice(eList, printTitle = TRUE, qUnit = 2, cex = 0.8,
  cex.main = 1.1, logScale = TRUE, cex.axis = 1.1, tcl = 0.5,
  las = 1, tinyPlot = FALSE, usgsStyle = FALSE, customPar = FALSE, ...)

Arguments

See Also

boxplot

Examples

eList <- Choptank_eList
# Water year:
boxQTwice(eList)
boxQTwice(eList, qUnit=1)
boxQTwice(eList, qUnit='cfs')
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
boxQTwice(eList)

A box plot of WRTDS residuals by month

Description

This function produces a boxplot of the residuals from WRTDS, expressed in natural log concentration units. It provides an alternative for viewing the standardized residuals, where the each residual is divided by its estimated standard error. The monthly boxplot widths are proportional to the square root of the sample size. The residuals for a censored value are determined as the difference between the natural log of the average of the upper and lower. bounds on the sample value, minus the log space estimate of concentration.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata

Usage

boxResidMonth(eList, stdResid = FALSE, las = 1, printTitle = TRUE,
  cex = 0.8, cex.axis = 1.1, cex.main = 1.1, font.main = 2,
  tinyPlot = FALSE, customPar = FALSE, monthLab = 1,
  randomCensored = FALSE, ...)

Arguments

See Also

boxplot

Examples

eList <- Choptank_eList
# Water year:
boxResidMonth(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart = 6, paLong = 3)
boxResidMonth(eList)

Calculates monthly mean values of Q, Conc, FNConc, Flux, and FNFlux for the entire record. If WRTDSKalman has been run it also includes the monthly mean values of GenConc and GenFlux.

Description

Computes the monthly mean values of discharge, concentration, flux, flow-normalized concentration and flow-normalized flux (Q, Conc, FNConc, Flux, and FNFlux) in SI units If WRTDSKalman has been run the outputs are averages for Q, Conc, GenConc, FNConc, Flux, GenFlux, and FNFlux. Note that the Flux, GenFlux, and FNFlux values are average flux values (not totals). For discharge the units are in m3/s, concentration is mg/L, and flux is kg/day. It returns a data frame containing month, year, decimal year, and mean values of DecYear, Q, Conc, GenConc, FNConc, Flux, GenFlux, and FNFlux.

Usage

calculateMonthlyResults(eList)

Arguments

Value

MonthlyResults data frame of numeric values describing the monthly average values

Examples

eList <- Choptank_eList
monthlyResults <- calculateMonthlyResults(eList)

Generic plotting function to create censored line segments

Description

Basic plotting framework for EGRET dot plots. Graphical parameters default to values that work well with most plots, but all can be re-assigned. See ?par for complete definitions of most optional input variables.

Usage

censoredSegments(yBottom, yLow, yHigh, x, Uncen, col = "black", lwd = 1)

Arguments

See Also

segments

Examples

x <- c(1,2,3,4,5,6)
y <- c(1,3,4,3.3,4.4,7)
xlim <- c(min(x)*.75,max(x)*1.25)
ylim <- c(0,1.25*max(y))
xlab <- "Date"
ylab <- "Concentration"
xTicks <- pretty(xlim)
yTicks <- pretty(ylim)
genericEGRETDotPlot(x=x, y=y, 
                    xlim=xlim, ylim=ylim,
                    xlab=xlab, ylab=ylab,
                    xTicks=xTicks, yTicks=yTicks,
                    plotTitle="Test"
)
yBottom <- 0
yLow <- c(NA,3,4,3.3,4,7)
yHigh <- c(1,3,4,3.3,5,NA)
Uncen <- c(0,1,1,1,0,0)
censoredSegments(yBottom=yBottom,yLow=yLow,yHigh=yHigh,x=x,Uncen=Uncen)

checkStartEndDate

Description

Checks that the start date is before the end date. If not, it will give the user the opportunity to correct, otherwise will create a warning.

Usage

checkStartEndDate(startDate, endDate, interactive = TRUE)

Arguments

Value

vector where first value is startDate, second is endDate

Examples

	
startDate <- '1985-01-01'		
endDate <- '1990-01-01'		
checkStartEndDate(startDate, endDate)

checkSurfaceSpan

Description

checkSurfaceSpan

Usage

checkSurfaceSpan(eList)

Arguments

Examples

 
eList <- Choptank_eList
checkSurfaceSpan(eList)

cleanUp eList

Description

Takes an eList as the input. If there are duplicated dates in the Sample data frame, will randomly select one value for that date. If there are censored values in the data set they will be replaced by random censored values. If there are no days with duplicate samples and no censored valued then the eList returned by the function will be identical to the eList that is passed to it.

Usage

cleanUp(eList, seed = NA)

Arguments

Details

This function is run before each iteration of generating a random sequence in the WRTDSKalman function

Value

eList with duplicated dates in the Sample data frame randomly sampled and censored values are replaced by random values.

Examples

eList <- Choptank_eList

eList <- cleanUp(eList)

Compress sample data frame

Description

Using raw data that has at least dateTime, value, code, populates the measured data portion of the Sample dataframe used in EGRET. ConcLow = Lower bound for an observed concentration ConcHigh = Upper bound for an observed concentration Uncen = 1 if uncensored, 0 if censored

Usage

compressData(data, verbose = TRUE)

Arguments

Value

data frame returnDataFrame data frame containing dateTime, ConcHigh, ConcLow, Uncen

Examples

dateTime <- c('1985-01-01', '1985-01-02', '1985-01-03')
comment1 <- c("","","")
value1 <- c(1,2,3)
comment2 <- c("","<","")
value2 <- c(2,3,4)
comment3 <- c("","","<")
value3 <- c(3,4,5)
dataInput <- data.frame(dateTime, comment1, value1, 
      comment2, value2, 
      comment3, value3, stringsAsFactors=FALSE)
compressData(dataInput)

concUnit class

Description

Some details about the concUnit class

Details

longPrefix

A character specifying the long name for concentration labels.

shortPrefix

A character specifying the short name for concentration labels.

Cumulative flow calculation

Description

This function computes the first day of the calendar year at which a specific fraction of the cumulative flow for that year has been exceeded. Typically one looks for the point where half the cumulative flow has happened (fract = 0.5). The portion of the year being considered is set by paStart and paLong. The matrix returned has 2 columns: the first is the year (integer when the period of analysis ends), the second is the day of the year when the fraction has been exceeded. None of the rows will have any NA values.

Usage

cumQdate(eList, paStart = 10, paLong = 12, fract = 0.5)

Arguments

Details

It is common to use this type of analysis on the snowmelt period of the year. If (for example) we assume that snowmelt starts with the month of March and ends in July then we would set paStart = 3 and paLong = 5

Value

annualSeries an integer matrix of two columns. The first column is the calendar year for the end of the period The second column is day of the year when the flow has exceeded the specified fraction of the entire period being considered

Examples

eList <- Choptank_eList
annualFlow <- cumQdate(eList)
head(annualFlow)
plot(annualFlow)
mod1 <- stats::lm(annualFlow[,2] ~ annualFlow[,1])
summary(mod1)

Data Overview for WRTDS

Description

Gives a summary of data to be used for WRTDS analysis

Usage

dataOverview(Daily, Sample)

Arguments

See Also

mergeReport

Examples

eList <- Choptank_eList
exDaily <- getDaily(eList)
exSample <- getSample(eList)
dataOverview(Daily = exDaily, Sample = exSample)

Check date format

Description

Checks to see if format is YYYY-MM-DD. Also performs a few other date checks.

Usage

dateFormatCheck(date)

Arguments

Value

condition logical TRUE or FALSE if checks passed or failed

Examples

date <- '1985-01-01'
dateFormatCheck(date)
dateWrong <- '1999/1/7'
dateFormatCheck(dateWrong)

decimalDate

Description

Create a decimal date or date/time from a vector.

Usage

decimalDate(rawData)

Arguments

Examples

dateTime <- c('1984-02-28 13:56',
              '1984-03-01 00:00',
              '1986-03-01 00:00',
              '1986-10-15 00:00')
decimalDate(dateTime)

dateTime <- c('1984-02-28', 
              '1984-03-01',
              '1986-03-01',
              '1986-10-15')
decimalDate(dateTime)

decimalHighLow

Description

decimalHighLow figures out the highest and lowest decimal year based on water year. The input is a data frame with columns Month and DecYear.

Usage

decimalHighLow(df)

Arguments

Value

list with DecHigh and DecLow (water year high/low decimal values)

Examples

eList <- Choptank_eList
highLow <- decimalHighLow(eList$Sample)

DecHigh <- highLow[["DecHigh"]]
DecLow <- highLow[["DecLow"]]

Error statistics

Description

This function takes a fitted WRTDS model and computes error statistics the residuals used here are cross-validation residuals, which will be slightly larger than regular regression residuals in the case of censored data, the residuals are computed from random residuals computed from makeAugmentedSample(), the function returns a list of error statistics and also prints them to the console

Usage

errorStats(eList)

Arguments

Value

erStats a numeric vector consisting of the following statistics RsqLC the R squared value for predictions of ln(Concentration) RsqLF the R squared value for predictions of ln(Flux) rmse the root mean squared error for ln(Concentration), same value would apply for Flux sepPercent the standard error of prediction for Concentration, expressed in percent same value would apply for Flux

Examples

eList <- Choptank_eList
errorStats(eList)

Jack-Knife cross validation of the WRTDS (Weighted Regressions on Time, Discharge, and Season)

Description

This function fits the WRTDS model n times (where n is the number of observations). For each fit, the data value being estimated is eliminated from the record. This gives predictions that do not depend on knowing the actual result for that day. Thus it provides for a more "honest" estimate of model performance than a traditional error analysis that uses all the data.

Usage

estCrossVal(DecLow, DecHigh, Sample, windowY = 7, windowQ = 2,
  windowS = 0.5, minNumObs = 100, minNumUncen = 50, edgeAdjust = TRUE,
  verbose = TRUE)

Arguments

Value

SampleCrossV data frame containing the sample data augmented by the results of the cross-validation exercise

Examples

eList <- Choptank_eList
Sample <- getSample(eList)
Daily <- getDaily(eList)
numDays <- length(Daily$DecYear)
DecLow <- Daily$DecYear[1]
DecHigh <- Daily$DecYear[numDays]

SampleCrossV <- estCrossVal(DecLow,DecHigh,Sample)

Estimates all daily values of Concentration, Flux, Flow-Normalized Concentration, and Flow Normalized Flux

Description

Uses the surfaces matrix estimated in estSurfaces to estimate 6 daily time series and appends them to the Daily data frame. The time series are (in order): yHat, the estimated natural log of concentration, dimensionless SE, the standard error of the natural log of concentration ConcDay, the estimated concentration in mg/L FluxDay, the estimated flux in kg/day FNConc, the flow-normalized concentration in mg/L FNFlux, the flow-normalized flux in kg/day

Bin the LogQ values by day-of-year.

Usage

estDailyFromSurfaces(eList, localsurfaces = NA, localDaily = NA)

getConcFluxFromSurface(eList, allLogQsByDayOfYear, localDaily,
  localsurfaces = NA)

getSurfaceEstimates(eList, localsurfaces = NA, localDaily = NA)

bin_Qs(localDaily)

Arguments

Details

The results are stored in an augmented version of the Daily data frame, which is returned as part of an EGRET object.

Value

egret object with altered Daily dataframe

Daily dataframe with yHat, SE, ConcDay and FluxDay calulated

Examples

eList <- Choptank_eList
#################################################
# This is usually done in modelEstimation:
Daily <- getDaily(eList)
surfaceIndexParameters<-surfaceIndex(Daily)
INFO <- eList$INFO
INFO$bottomLogQ<-surfaceIndexParameters[['bottomLogQ']]
INFO$stepLogQ<-surfaceIndexParameters[['stepLogQ']]
INFO$nVectorLogQ<-surfaceIndexParameters[['nVectorLogQ']]
INFO$bottomYear<-surfaceIndexParameters[['bottomYear']]
INFO$stepYear<-surfaceIndexParameters[['stepYear']]
INFO$nVectorYear<-surfaceIndexParameters[['nVectorYear']]
eList$INFO <- INFO
#################################################

Daily <- estDailyFromSurfaces(eList)

Estimate the three surfaces (for yHat, SE and ConcHat) as a function of DecYear and logQ and store in the three-dimensional object called surfaces

Description

This function uses weighted survival regression to estimate three surfaces that cover the complete range of DecYear and log(Q) values in the Daily data set. These surfaces are: (1) is the estimated log concentration (yHat), (2) is the estimated standard error (SE), (3) is the estimated concentration (ConcHat). They are mapped as an array that covers the complete space of daily discharge and time. The first index is discharge, layed out in 14 equally spaced levels of log(Q). The second index is time, layed out as 16 increments of the calendar year, starting January 1. It returns the 3 dimensional array called surfaces. This array will be used to estimate these 3 quantities for any given day in the daily values record.

Usage

estSurfaces(eList, surfaceStart = NA, surfaceEnd = NA, localSample = NA,
  windowY = 7, windowQ = 2, windowS = 0.5, minNumObs = 100,
  minNumUncen = 50, edgeAdjust = TRUE, verbose = TRUE,
  interactive = NULL, run.parallel = FALSE)

Arguments

Value

surfaces array containing the three surfaces estimated, array is 3 dimensional

Examples

eList <- Choptank_eList

surfaces <- estSurfaces(eList)

surfaceStart <- "1984-10-01"
surfaceEnd <- "1986-09-30"
surfaces_1 <- estSurfaces(eList, surfaceStart, surfaceEnd)

wall_sample <- head(eList$Sample, n=500)

surface_wall <- estSurfaces(eList, localSample = wall_sample)

Update Sample dataframe

Description

Used for updating the Sample dataframe if ConcLow or ConcHigh is manually adjusted. Adjusts ConcAve and Uncen columns.

Usage

fixSampleFrame(eList)

Arguments

Value

localSample data frame

Examples

eList <- Choptank_eList
Sample <- eList$Sample
Sample[1,c("ConcLow","ConcHigh")] <- c(NA, 0.01) # Adjusted to left-censored
Sample[2,c("ConcLow","ConcHigh")] <- c(1.1, 1.3) # Adjusted to interval-censored
Sample[3,c("ConcLow","ConcHigh")] <- c(1.3, 1.3) # Simple adjustment
eList$Sample <- Sample
eList <- fixSampleFrame(eList)
eList$Sample[1:3,]

Flexible Flow Normalization

Description

This function implements generalized flow normalization. This means that for determining the flow normalized concentration and flow normalized flux for any given year, there is a specified list of years from which to create the discharge record used in the flow-normalization process. That set of years is defined by the dateInfo object.

Usage

flexFN(eList, dateInfo, localsurfaces = NA, oldSurface = FALSE,
  flowNormStartCol = "flowNormStart", flowNormEndCol = "flowNormEnd",
  flowStartCol = "flowStart", flowEndCol = "flowEnd")

Arguments

Value

named list, eList, containing INFO, Daily, Sample, and surfaces objects

Examples

eList <- Choptank_eList
eList <- setUpEstimation(eList)
flowNormStart <- c("1979-10-01","1990-01-01","1992-10-10")
flowNormEnd <- c("1995-06-06","2004-03-03","2011-09-29")
flowStart <- c("1979-10-01","1995-06-07","2004-03-04")
flowEnd <- c("1995-06-06","2004-03-03","2011-09-29") 
dateInfo <- data.frame(flowNormStart,
                       flowNormEnd,
                       flowStart, 
                       flowEnd, 
                       stringsAsFactors = FALSE)

newEList <- flexFN(eList, dateInfo)
plotFluxHist(newEList)
flexPlotAddOn(newEList)

wallSurface <- estSurfaces(eList, localSample = eList$Sample[1:500,])
wallEList <- flexFN(eList, dateInfo, localsurface = wallSurface)
plotFluxHist(wallEList)

Flexible Flow Normalization Plot Add On

Description

Flexible Flow Normalization Plot Add On

Usage

flexPlotAddOn(eList, showArrows = TRUE, showRect = TRUE,
  customPalette = NULL)

Arguments

Examples

eList <- Choptank_eList
eList <- setUpEstimation(eList)
flowNormStart <- c("1979-10-01","1990-01-01","1992-10-10")
flowNormEnd <- c("1995-06-06","2004-03-03","2011-09-29")
flowStart <- c("1979-10-01","1995-06-07","2004-03-04")
flowEnd <- c("1995-06-06","2004-03-03","2011-09-29") 
dateInfo <- data.frame(flowNormStart,
                       flowNormEnd,
                       flowStart, 
                       flowEnd, 
                       stringsAsFactors = FALSE)

newEList <- flexFN(eList, dateInfo)
plotFluxHist(newEList)
flexPlotAddOn(newEList)

plotFluxHist(newEList)
flexPlotAddOn(newEList, customPalette=c("#d5ce48", "#fd300f", "#3e0289"))

Computes several values of the flow duration curve for streamflow centered on a specific date of the year

Description

This function is useful for helping the analyst determine the empirical probability distribution of streamflow for a particular part of the year or for the whole year. This is particularly useful in setting up discharge scales for various other plots in this package.

Usage

flowDuration(eList, centerDate = "09-30", qUnit = 2, span = 365,
  monthLab = 1)

Arguments

Value

qDuration A named vector with flow duration information.

Examples

eList <- Choptank_eList
# for a window of 30 days either side of June 25 expressed in units
# of cfs:
flowDuration(eList, "06-25", qUnit = 1, span = 30) 
# for a flow-duration curve covering the whole year, 
# expressed in units of cms, and returning a data frame of results: 
qDuration <- flowDuration(eList, qUnit = 2) 

Produces 8-panel plot that is useful for determining if there is a flux bias problem

Description

These plots use the jack-knife estimates from WRTDS to investigate the potential flux bias problem. It can also be used for estimates constructed by other methods (such as LOADEST) if the results are stored in a data frame organized like the Sample data frame. It allows additional label information to indicate what method is used. The use of this plot is described in Hirsch, Robert M., 2014. Large Biases in Regression-Based Constituent Flux Estimates: Causes and Diagnostic Tools. Journal of the American Water Resources Association (JAWRA) 1-24. DOI: 10.1111/jawr.12195

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

fluxBiasMulti(eList, qUnit = 2, fluxUnit = 3, moreTitle = "WRTDS",
  cex = 0.7, cex.axis = 1.1, cex.main = 1.1, randomCensored = FALSE,
  col = "black", lwd = 1, concLab = 1, monthLab = 1, ...)

Arguments

Examples

eList <- Choptank_eList
# Water year:
fluxBiasMulti(eList)
fluxBiasMulti(eList, fluxUnit = 2)
# Graphs consisting of Jun-Aug
eList <- setPA(eList,paStart=6,paLong=3)
fluxBiasMulti(eList)

Compute the flux bias statistic: (mean of estimated flux - mean of observed flux) / mean of estimated flux

Description

Computes three versions of the flux bias: The first where all censored values are set to their miniumum. The second where all censored values are set to their maximum. The third which is the average of the other two. In practice there is rarely a noticable difference among them.

Usage

fluxBiasStat(localSample)

Arguments

Value

fluxBias a vector of three numerical values, a lower bound, upper bound and an average estimate of the ratio of (mean estimated flux - mean observed flux) / mean estimated flux. Typically one should use fluxBias[3]

Examples

eList <- Choptank_eList
Sample <- getSample(eList)
fluxBias <- fluxBiasStat(Sample) 

fluxUnit class

Description

Some details about the fluxUnit class

Details

shortName

A character specifying the short name.

unitFactor

A numeric representing the conversion factor

unitName

A character specifying the full name.

unitExpress

An expression specifying the full name starting with Observed.

unitExpressTiny

An expression specifying the abbreviated name starting with Observed.

unitEstimate

An expression specifying the full name starting with Estimated.

unitEstimateTiny

An expression specifying the abbreviated name starting with Estimated.

unitUSGS

A character specifying flux with full text.

shortCode

A number for quick lookup

Axis generation for log discharge

Description

Discharge axis tick generation

Usage

generalAxis(x, maxVal, minVal, units = NA, logScale = FALSE,
  tinyPlot = FALSE, padPercent = 5, concentration = TRUE, concLab = 1,
  usgsStyle = FALSE, prettyDate = TRUE)

Arguments

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)
INFO <- getInfo(eList)
x <- Daily$Q
max <- max(x)
min <- 0
units <- INFO$param.units
generalAxis(x, max, min, units)
min <- min(x)
generalAxis(x, max, min, units, log=TRUE)
generalAxis(Daily$ConcDay, 100, 0, concLab = "concentration")

Generic EGRET plotting function

Description

Basic plotting framework for EGRET dot plots. Graphical parameters default to values that work well with most plots, but all can be re-assigned. See ?par for complete definitions of most optional input variables.

Usage

genericEGRETDotPlot(x, y, xlim, ylim, xTicks = pretty(xlim),
  yTicks = pretty(ylim), printTitle = TRUE, xaxs = "i", xlab = "",
  yaxs = "i", ylab = "", plotTitle = "", pch = 20, cex = 0.7,
  cex.main = 1.3, font.main = 2, cex.lab = 1.2, tcl = 0.5,
  cex.axis = 1, las = 1, xDate = FALSE, tinyPlot = FALSE,
  hLine = FALSE, oneToOneLine = FALSE, rmSciX = FALSE, rmSciY = FALSE,
  customPar = FALSE, col = "black", lwd = 1, showXLabels = TRUE,
  showYLabels = TRUE, showXAxis = TRUE, showYAxis = TRUE,
  removeFirstX = FALSE, removeLastX = FALSE, removeFirstY = FALSE,
  removeLastY = FALSE, ...)

Arguments

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)
x <- Daily$Date
y <- Daily$Q
xlim <- c(min(x),max(x))
ylim <- c(min(y),1.05*max(y))
xlab <- "Date"
ylab <- "Flow"
genericEGRETDotPlot(x=x, y=y, 
                    xlim=xlim, ylim=ylim,
                    xlab=xlab, ylab=ylab,
                    plotTitle="Test"
)

genmissing

Description

Generates a lag one auto-regressive time series, where the first and last values are fixed. Marginal expected value is zero and variance is one. Generated values have a normal conditional distribution.

Usage

genmissing(X1, XN, rho, N)

Arguments

Value

genmissing numeric vector of length N, conditioned on the first value (X1) and last value (XN) with the specified lag one autocorrelation in the limit (where N is large) the values are normal with mean 0 and variance 1

Author(s)

Tim Cohn

Get Daily dataframe from EGRET object

Description

From a named list or EGRET object, extract the Daily dataframe

Usage

getDaily(x, ...)

## S3 method for class 'egret'
getDaily(x, ...)

## Default S3 method:
getDaily(x, ...)

Arguments

Value

Daily dataframe

See Also

readNWISDaily, readNWISSample

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)

Get INFO dataframe from EGRET object

Description

From a named list or EGRET object, extract the INFO dataframe

Usage

getInfo(x, ...)

## S3 method for class 'egret'
getInfo(x, ...)

## Default S3 method:
getInfo(x, ...)

Arguments

Value

INFO dataframe

See Also

readNWISDaily, readNWISSample

Examples

eList <- Choptank_eList
INFO <- getInfo(eList)

Get Sample dataframe from EGRET object

Description

From a named list or EGRET object, extract the Sample dataframe

Usage

getSample(x, ...)

getSample(x, ...)

getSample.default(x, ...)

Arguments

Value

Sample dataframe

See Also

readNWISDaily, readNWISSample

Examples

eList <- Choptank_eList
Sample <- getSample(eList)

Get surfaces matrix from EGRET object

Description

From a named list or EGRET object, extract the surfaces matrix

Usage

getSurfaces(x, ...)

## S3 method for class 'egret'
getSurfaces(x, ...)

## Default S3 method:
getSurfaces(x, ...)

Arguments

Value

Sample dataframe

See Also

readNWISDaily, readNWISSample

Examples

eList <- Choptank_eList
surfaces <- getSurfaces(eList)

Check for EGRET object

Description

Checks object to see if it is an EGRET object

Usage

is.egret(x)

Arguments

Value

logical

Examples

eList <- Choptank_eList
is.egret(eList)

jitter Sample

Description

This function is used in cases where there are numerical problems with the estimation of the WRTDS model. This mostly happens during bootstrap estimation or when the data sets are very large. In order to reduce the collinearity in the explanatory variables, some random noise is added to the time and log discharge variables in the Sample data frame.

Usage

jitterSam(Sam, V = 0.2)

Arguments

Value

SamR a data frame structured like the Sam data frame but with the time and discharge variables modified by adding random jitter

Examples

eList <- Choptank_eList
Sample_jitter <- jitterSam(eList$Sample)

Sets up tick marks for an axis with a log scale, where the graph is small

Description

Axis tick marks for a log scale for cases where the data cover many orders of magnitude and the graph is small. These tick marks are designed to progress by factors of 10.

Usage

logPretty1(xMin, xMax)

Arguments

Value

xTicks A vector representing the values for each of the tick marks

Examples

xMin <- 0.7
xMax <- 990000
logPretty1(xMin, xMax)
xMin <- 3
xMax <- 15
logPretty1(xMin, xMax)

Sets up tick marks for an axis with a log scale

Description

Axis tick marks for a log scale. These tick marks are designed to progress with 3 tick marks for every factor of 10. For example: 2,5,10,20,50,100,200,500.

Usage

logPretty3(xMin, xMax)

Arguments

Value

xTicks A vector representing the values for each of the tick marks

Examples

logPretty3(0.7, 990000)
logPretty3(3, 15)

Produces annual series of 8 streamflow statistics (and a lowess smooth of them) from daily streamflow data

Description

Part of the flowHistory system. The data come from Daily and INFO data frames. Note that the function setPA must be run before this to establish the period of analysis (e.g. water year).

Usage

makeAnnualSeries(eList, edgeAdjust = TRUE)

Arguments

Details

The smooth is a loess smooth computed on the log flow values and then transformed back to real space Smoothing window is a fixed number of years, the window width has a default value of 20 years but can be modified by changing its value in INFO data frame (using setPA function)

Value

annualSeries matrix that contains the annual series of streamflow statistics annualSeries is a matrix 3 * 8 * numYears, where numYears is the number of years in the data set in the first dimension 1 is the year, 2 is the actual value, 3 is the smoothed value in the second dimension, the index is the istat value (identifying the flow statistic) the third dimension is year

Examples

eList <- Choptank_eList
annualSeries <- makeAnnualSeries(eList)

Create randomized residuals and observations for data sets that have some censored data

Description

This function is used to add two columns to the Sample data frame: rResid and rObserved. rResid is the randomized residual value computed in log concentration units, and rObserved is the randomized 'observed' value of concentration in concentration units. Both of these are computed for all censored samples ("less than values"). They are created for purposes of plotting and are not used in any computations in EGRET.

Usage

makeAugmentedSample(eList)

Arguments

Details

The WRTDS model must be estimated before this function can be run. The random value that is generated lies between the reporting limit and zero and is distributed as a truncated log-normal distribution, with parameters derived from the fitted WRTDS model. These random values are never used in any computations in EGRET but are used for purposes of plotting the data set or residuals. When plotted in other functions they are shown as open circles.

Value

eList named list with modified Sample data frame.

Examples

choptankAugmented <- makeAugmentedSample(Choptank_eList)

makeDateInfo

Description

Create a data frame that organizes date segmentations for runSeries.

Usage

makeDateInfo(windowSide, surfaceStart, surfaceEnd, firstQDate0, lastQDate0)

Arguments

Examples

windowSide <- 7
surfaceStart <- "1984-01-01"
surfaceEnd <- "2012-12-31"
firstQDate0 <- "1970-01-01"
lastQDate0 <- "2014-06-01"
dateInfo <- makeDateInfo(windowSide, 
                         surfaceStart, surfaceEnd, 
                         firstQDate0, lastQDate0)

mergeReport

Description

This function does three things. 1) It transfers the daily discharge value from the Daily data frame to to Sample data frame for those days with samples. 2) It merges the INFO, Daily and Sample data frames to form an eList object, 3) and it prints out a "report" of basic information about the Daily and Sample data frames.

Usage

mergeReport(INFO, Daily, Sample = NA, surfaces = NA, verbose = TRUE,
  interactive = NULL)

Arguments

Details

There must be an INFO and a Daily data frame for this function to work. That would be the case for a study of flow only, with no consideration of water quality. If water quality is being considered then INFO, Daily, and Sample all need to be provided in the call to this function.

Note that the Sample dataframe in the global environment does not update with the flow information.

Value

eList named list with at least INFO, and Daily data frames. It can also include a Sample data frame.

See Also

readNWISDaily, readNWISSample

Examples

siteNumber <- '01491000'
pCode <- '00631'

Daily <- readNWISDaily(siteNumber,'00060', '1984-10-01', '')
Sample <- readNWISSample(siteNumber,pCode, '1984-10-01', '')
INFO <- readNWISInfo(siteNumber,pCode,interactive=FALSE)
eList <- mergeReport(INFO, Daily, Sample)
Sample <- eList$Sample
plot(eList)

# Create eList with no water quality data:

eList <- mergeReport(INFO, Daily, Sample = NA)
plotFour(eList)

Estimation process for the WRTDS (Weighted Regressions on Time, Discharge, and Season)

Description

This one function does three things. 1) a jack-knife cross-validation of a WRTDS model in which it augments the Sample data frame in the eList, 2) fits the WRTDS model creating the surfaces matrix and places it in the eList (the surfaces matrix expresses the estimated concentration as a function of discharge and time), and 3) estimates the daily values of concentration and flux, and flow normalized concentration and flux and places these in the Daily data frame in the eList. It returns a named list with the following dataframes: Daily, INFO, Sample, and the matrix: surfaces.

Usage

modelEstimation(eList, windowY = 7, windowQ = 2, windowS = 0.5,
  minNumObs = 100, minNumUncen = 50, edgeAdjust = TRUE, verbose = TRUE,
  run.parallel = FALSE)

Arguments

Value

eList named list with INFO, Daily, and Sample dataframes, along with the surfaces matrix.

Examples

eList <- Choptank_eList

eList <- modelEstimation(eList)

monthLabel class

Description

Some details about the monthLabel class

Details

monthAbbrev

A character specifying the abbreviated month name.

monthFull

A character specifying the full month name

monthSingle

A character specifying the single letter of the month.

Produces a 4 panel plot that gives an overview of the data set prior to any processing

Description

This function produces the 4 plots based only on the data stored in the eList. The four plots are 1) log concentration versus log discharge, 2) log concentration versus time 3) a boxplot of log concentration by month, and 4) a side-by-side boxplot of the sampled discharges and all daily discharges. To save space, the graphic is labeled only at the top of the 4 graph display.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

multiPlotDataOverview(eList, qUnit = 2, cex.main = 1.2,
  randomCensored = FALSE, logScaleConc = TRUE, logScaleQ = TRUE,
  concLab = 1)

Arguments

See Also

plotConcQ, boxConcMonth, plotConcTime, boxQTwice

Examples

eList <- Choptank_eList
# Water year:
multiPlotDataOverview(eList, qUnit=1)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
multiPlotDataOverview(eList, qUnit=1) 

# Custom axes:
eList$INFO$param.units <- "ng"
qConst_precip <- new("qUnit",
                     qShortName = "   mm  ",
                     qUnitFactor = 1,
                     qUnitName = "Millimeter",
                     qUnitExpress = expression(paste("Precipitation in ",mm)),
                     qUnitTiny = expression(paste("Precipitation ", "(", mm, ")")),
                     shortCode = 1,
                     unitUSGS = "Precipitation, in mm",
                     prefix = "Precipitation")

deposition <- new("concUnit",
                  longPrefix = "Deposition",
                  shortPrefix = "Dep")

multiPlotDataOverview(eList, 
                      qUnit = qConst_precip, 
                      concLab = deposition)

Makes 15 graphs of streamflow statistics on a single page. These encompass the 7-day minimum, mean, and 1-day maximum for each of the following 5 Periods of Analysis: Annual, Fall, Winter, Spring and Summer.

Description

Part of flowHistory system. All results are expressed as runoff (mm/day). The individual plots are constructed by the same method as used in plotFlowSingle. The annual results are based on the Water Year. The seasons are defined as the following groups of months: SON, DJF, MAM, JJA.

Usage

plot15(eList, yearStart, yearEnd)

Arguments

Details

For formatting purposes it is best to use the following commands before calling the plot15 function (savePath is the pathname for directory to store the output) # plotName <- paste(savePath, "plot15.", eList$INFO$shortName, ".ps", sep = "") # postscript(file = plotName, width = 8, height = 10, horizontal = FALSE, family = "Helvetica") Then after running plot15, the user needs to give the command dev.off()

See Also

plot1of15

Examples

eList <- Choptank_eList

plot15(eList, yearStart = 1980, yearEnd = 2010)
dev.off()

plots 1 of the 15 graphs of streamflow statistics on a single page

Description

Part of the flowHistory system. It is designed to create each of the component graphs for the function plot15. The 15 graphs include annual and four seasonal graphs for each of 3 flow statistics: 1-day maximum, mean, and 7-day minimum. The computations involved are the same as the ones used in plotFlowSingle or in makeAnnualSeries

Usage

plot1of15(eList, yearStart, yearEnd, qf, istat, isBottom = FALSE)

Arguments

Examples

eList <- Choptank_eList
plot1of15(eList, 1980, 2010, 0.2938476, 5)

Graph of annual concentration and flow normalized concentration versus year

Description

Data come from named list (eList), which contains a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

The annual concentrations are "time-weighted" mean concentrations (as opposed to "flow-weighted"). The annual results reported are for a specified "period of analysis" which can be an entire water year, a calendar, a season or even an individual month. User specifies this period of analysis in the call to setupYears.

User can specify plotting of three possible series. All are in units of mg/L. Annual mean concentration WRTDS_K version of annual mean concentration (requires that WRTDSKalman has been run) Flow normalized mean concentration

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

plotConcHist(eList, yearStart = NA, yearEnd = NA, concMax = NA,
  printTitle = TRUE, tinyPlot = FALSE, usgsStyle = FALSE,
  plotFlowNorm = TRUE, plotAnnual = TRUE, plotGenConc = FALSE,
  cex = 0.8, cex.axis = 1.1, cex.main = 1.1, lwd = 2, col = "black",
  col.pred = "green", concLab = 1, col.gen = "red", customPar = FALSE,
  ...)

Arguments

See Also

setupYears, genericEGRETDotPlot

Examples

yearStart <- 2001
yearEnd <- 2010
eList <- Choptank_eList

plotConcHist(eList, yearStart, yearEnd)

Plot of Observed Concentration versus Estimated Concentration

Description

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

plotConcPred(eList, concMax = NA, logScale = FALSE, printTitle = TRUE,
  tinyPlot = FALSE, cex = 0.8, cex.axis = 1.1, cex.main = 1.1,
  customPar = FALSE, col = "black", lwd = 1, randomCensored = FALSE,
  concLab = 1, usgsStyle = FALSE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotConcPred(eList)

# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotConcPred(eList)

Plot of Observed Concentration versus Discharge

Description

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata. Discharge is plotted on a log scale.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

plotConcQ(eList, qUnit = 2, tinyPlot = FALSE, logScale = FALSE,
  randomCensored = FALSE, concMax = NA, concMin = NA,
  printTitle = TRUE, cex = 0.8, cex.axis = 1.1, cex.main = 1.1,
  usgsStyle = FALSE, rmSciX = FALSE, rmSciY = FALSE, customPar = FALSE,
  col = "black", lwd = 1, concLab = 1, ...)

Arguments

Details

The function has two possible ways to plot censored values (e.g. "less-than values").

The default is to plot them as a vertical line that goes from the reporting limit down to the bottom of the graph.

The alternative is to set randomCensored = TRUE. In this case a random value is used for plotting the individual sample value. This random value lies between the reporting limit and zero and it is distributed as a truncated log normal based on the fitted WRTDS model.

The function makeAugmentedSample must be run first if randomCensored = TRUE. Running makeAugmentedSample requires that modelEstimation has already been run.

These random censored values are used to create more readable plots and are not used in any computations about the data set. The random censored values are shown as open circles and the non-censored data are shown as filled dots.

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotConcQ(eList)
plotConcQ(eList, logScale=TRUE)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotConcQ(eList)

Plot up to three curves representing the concentration versus discharge relationship. Each curve is a different point in time.

Description

These plots are like a vertical slice of the estimated concentration surface that is seen in the plotContours function. These plots show how the concentration-discharge relationship is changing over time. Typically the time points selected would be in three years at the same time of year spaced out over the period of record. But that is not necessary. Another possibility is to use this to explore seasonal differences. In this case the three dates would be in the same year but different times during the year.

This plot can also help identify situations where the windowQ may be too small. If there are substantial oscillations of some of the curves, then the windowQ should be increased. Alternatively, windowQ may be too large. This can be seen when the windowQ is reduced (say to 1.0). A good choice of windowQ would be a value just great enough to damp out oscillations in the curves.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Usage

plotConcQSmooth(eList, date1, date2, date3, qLow, qHigh, qUnit = 2,
  legendLeft = 0, legendTop = 0, concMax = NA, concMin = NA,
  bw = FALSE, printTitle = TRUE, printValues = FALSE, minNumObs = 100,
  minNumUncen = 50, colors = c("black", "red", "green"),
  printLegend = TRUE, windowY = 7, windowQ = 2, windowS = 0.5,
  tinyPlot = FALSE, customPar = FALSE, lwd = 2, cex = 0.8,
  cex.axis = 1.1, cex.main = 1.1, cex.legend = 1.2, lineVal = c(1, 1,
  1), logScale = FALSE, edgeAdjust = TRUE, concLab = 1,
  usgsStyle = FALSE, ...)

Arguments

See Also

genericEGRETDotPlot, runSurvReg

Examples

date1 <- "1982-06-01"
date2 <- "1994-06-01"
date3 <- "2010-06-01"
qLow <- 0.5
qHigh <- 50
eList <- Choptank_eList

plotConcQSmooth(eList, date1, date2, date3, qLow, qHigh,
                 legendLeft = 0.6, legendTop = 0.7)
plotConcQSmooth(eList, date1, date2, date3, qLow, qHigh,
                logScale=TRUE, legendLeft = 0.6, legendTop = 0.7)

Plot of Observed Concentration versus Time

Description

This function allows the user to plot all of the data, but also to limit it in two ways. The data can be limited to only those observed concentrations collected in a specified discharge range. The data can also be limited to only those observed in certain months of the year. These two selection criteria can be combined. For example, we may only want to plot data for discharges between 100 and 500 cubic feet per second in the months of March, April and May.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Usage

plotConcTime(eList, qUnit = 2, yearStart = NA, yearEnd = NA,
  qLower = NA, qUpper = NA, randomCensored = FALSE, tinyPlot = FALSE,
  concMax = NA, concMin = NA, printTitle = TRUE, logScale = FALSE,
  cex = 0.8, cex.axis = 1.1, cex.main = 1.1, customPar = FALSE,
  col = "black", lwd = 1, concLab = 1, usgsStyle = FALSE, ...)

Arguments

Details

The function has two possible ways to plot censored values (e.g. "less-than values").

The default is to plot them as a vertical line that goes from the reporting limit down to the bottom of the graph.

The alternative is to set randomCensored = TRUE. In this case a random value is used for plotting the individual sample value. This random value lies between the reporting limit and zero and it is distributed as a truncated log normal based on the fitted WRTDS model.

The function makeAugmentedSample must be run first if randomCensored = TRUE. Running makeAugmentedSample requires that modelEstimation has already been run.

These random censored values are used to create more readable plots and are not used in any computations about the data set. The random censored values are shown as open circles and the non-censored data are shown as filled dots.

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotConcTime(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotConcTime(eList, 
             qUnit = 1,
             qLower = 100,
             qUpper = 10000)
plotConcTime(eList, logScale=TRUE)
plotConcTime(eList, 
             qUnit = 1, 
             qLower = 100, qUpper = 10000,
              randomCensored = TRUE)

Plot of the time series of daily concentration estimates and the sample values for the days that were sampled

Description

This plot is useful for visual examination of the ability of the WRTDS, or other model, to fit the data, seen in a time-series perspective. The graph is most useful when it covers a period of just a few years and not the complete record but a complete record can be done by repeated use over a series of segments.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotConcTimeDaily(eList, yearStart = NA, yearEnd = NA, tinyPlot = FALSE,
  concMax = NA, printTitle = TRUE, plotGenConc = TRUE, cex = 0.8,
  cex.axis = 1.1, randomCensored = FALSE, cex.main = 1.1,
  customPar = FALSE, col = "black", lwd = 1, prettyDate = TRUE,
  usgsStyle = FALSE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotConcTimeDaily(eList)
plotConcTimeDaily(eList, 
                  yearStart = 1998,
                  yearEnd = 2001,
                  plotGenConc = FALSE)

Plot up to three curves representing the concentration versus time relationship, each curve representing a different flow.

Description

These plots show how the concentration-time relationship is changing over flow.

This plot can also help identify situations where the windowY may be too small. If there are substantial oscillations of some of the curves, then the windowY should be increased. Alternatively, windowY may be too large. This can be seen when the windowY is reduced (say to 4.0). A good choice of windowY would be a value just great enough to damp out oscillations in the curves.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data and an INFO dataframe with metadata.

Usage

plotConcTimeSmooth(eList, q1, q2, q3, centerDate, yearStart, yearEnd,
  qUnit = 2, legendLeft = 0, legendTop = 0, concMax = NA,
  concMin = NA, bw = FALSE, printTitle = TRUE, colors = c("black",
  "red", "green"), printValues = FALSE, tinyPlot = FALSE, concLab = 1,
  monthLab = 1, minNumObs = 100, minNumUncen = 50, windowY = 10,
  windowQ = 2, windowS = 0.5, cex.main = 1.1, lwd = 2,
  printLegend = TRUE, cex.legend = 1.2, cex = 0.8, cex.axis = 1.1,
  customPar = FALSE, lineVal = c(1, 1, 1), logScale = FALSE,
  edgeAdjust = TRUE, usgsStyle = FALSE, ...)

Arguments

See Also

genericEGRETDotPlot, runSurvReg

Examples

q1 <- 1
q2 <- 10
q3 <- 100
centerDate <- "07-01"
yearStart <- 1990
yearEnd <- 2010
eList <- Choptank_eList
plotConcTimeSmooth(eList, q1, q2,q3, centerDate, 
                   yearStart, yearEnd, legendLeft = 1997, 
                   legendTop = 0.44, cex.legend = 0.9)
plotConcTimeSmooth(eList, q1, q2,q3, centerDate, yearStart, 
                   yearEnd, logScale = TRUE, legendLeft = 1994, 
                   legendTop = 0.4, cex.legend = 0.9)

Color contour plot of the estimated surfaces as a function of discharge and time (surfaces include log concentration, standard error, and concentration)

Description

These plots are normally used for plotting the estimated concentration surface (whatSurface = 3) but can be used to explore the estimated surfaces for the log of concentration or for the standard error (in log space) which is what determines the bias correction. The plots are often more interpretable when the time limits for the plot are less than a decade. To explore changes over a long time period it is best to do this multiple times, for various time slices of 2 years (for example) or to use the function plotDiffContours.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Obtaining a plot that provides good insight it is useful to experiment with several of the arguments such as yearStart, yearEnd, qBottom, qTop, and contourLevels.

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotContours(eList, yearStart, yearEnd, qBottom = NA, qTop = NA,
  whatSurface = 3, qUnit = 2, contourLevels = NA, span = 60,
  pval = 0.05, printTitle = TRUE, vert1 = NA, vert2 = NA, horiz = NA,
  tcl = 0.03, flowDuration = TRUE, customPar = FALSE, yTicks = NA,
  tick.lwd = 1, usgsStyle = FALSE, lwd = 2, cex.main = 1,
  cex.axis = 1, concLab = 1,
  color.palette = grDevices::colorRampPalette(c("white", "gray", "blue",
  "red")), ...)

Arguments

Examples

yearStart <- 2002
yearEnd <- 2010
qBottom <- 0.5
qTop<- 20
clevel <- seq(0,2,0.25)
eList <- Choptank_eList
plotContours(eList, yearStart, yearEnd, qBottom, qTop, 
             contourLevels = clevel)  
plotContours(eList, yearStart, yearEnd, qBottom, qTop = 50, 
             contourLevels = clevel, flowDuration = FALSE) 
colors <- grDevices::colorRampPalette(c("white","black"))
plotContours(eList, yearStart, yearEnd, qBottom, qTop = 50, 
             contourLevels = clevel, color.palette = colors, 
             flowDuration = FALSE)

Plots the difference between two years from a contour plot created by plotContours

Description

These plots are normally used for plotting changes in the estimated concentration surface (whatSurface=3) but can be used to explore the changes in estimated surfaces for the log of concentration or for the standard error (in log space) which is what determines the bias correction.

The difference can be shown either as an arithmetic difference or as a percentage difference.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotDiffContours(eList, year0, year1, qBottom = NA, qTop = NA,
  maxDiff = NA, whatSurface = 3, tcl = 0.03, qUnit = 2, span = 60,
  pval = 0.05, printTitle = TRUE, plotPercent = FALSE, vert1 = NA,
  vert2 = NA, horiz = NA, flowDuration = TRUE, yTicks = NA,
  tick.lwd = 1, lwd = 2, cex.main = 0.95, cex.axis = 1,
  customPar = FALSE, usgsStyle = FALSE,
  color.palette = grDevices::colorRampPalette(c("blue", "white", "red")),
  concLab = 1, monthLab = 1, ...)

Arguments

Examples

year0 <- 1990
year1 <- 2009
qBottom <- 0.5
qTop <- 20
maxDiff<-0.5
eList <- Choptank_eList
plotDiffContours(eList, year0, year1, qBottom, qTop, maxDiff = 0.5)
plotDiffContours(eList, year0, year1, qBottom, qTop, maxDiff = 50, plotPercent = TRUE)

Creates a plot of a time series of a particular flow statistic and a loess smooth of that flow statistic

Description

A part of the flowHistory system. The index of the flow statistics is istat. These statistics are: (1) 1-day minimum, (2) 7-day minimum, (3) 30-day minimum, (4) median (5) mean, (6) 30-day maximum, (7) 7-day maximum, and (8) 1-day maximum

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotFlowSingle(eList, istat, yearStart = NA, yearEnd = NA, qMax = NA,
  printTitle = TRUE, tinyPlot = FALSE, customPar = FALSE,
  runoff = FALSE, qUnit = 1, printStaName = TRUE, printPA = TRUE,
  usgsStyle = FALSE, printIstat = TRUE, cex = 0.8, cex.axis = 1.1,
  cex.main = 1.1, lwd = 2, col = "black", ...)

Arguments

Details

The curve plotted on the graph is a loess smooth of the data. This smooth is computed on the logs of the data and then transformed back to plot. The width of the smoothing window is 20 years on either side of the year being plotted However, the window width can be adjusted using setPA function.

See Also

makeAnnualSeries, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotFlowSingle(eList, 1)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotFlowSingle(eList, 1)

Graph of annual flux and flow normalized flux versus year

Description

The annual results reported are for a specified "period of analysis" which can be an entire water year, a calendar, a season or even an individual month. The user specifies this period of analysis in the call to setupYears. Values plotted express a flux rate, such as thousand kg per year For a period of analysis that is less than a year, this does not equal the mass transported over the period of analysis

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list (eList), which contains a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotFluxHist(eList, yearStart = NA, yearEnd = NA, fluxUnit = 9,
  fluxMax = NA, printTitle = TRUE, usgsStyle = FALSE,
  plotFlowNorm = TRUE, plotAnnual = TRUE, plotGenFlux = FALSE,
  tinyPlot = FALSE, col = "black", col.pred = "green", col.gen = "red",
  cex = 0.8, cex.axis = 1.1, cex.main = 1.1, lwd = 2,
  customPar = FALSE, ...)

Arguments

See Also

setupYears

Examples

yearStart <- 2001
yearEnd <- 2010
eList <- Choptank_eList
# Water year:

plotFluxHist(eList)
plotFluxHist(eList, yearStart, yearEnd, fluxUnit = 1)
plotFluxHist(eList, yearStart, yearEnd, fluxUnit = 'kgDay')

Graph of observed versus estimated flux

Description

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Usage

plotFluxPred(eList, fluxUnit = 3, fluxMax = NA, printTitle = TRUE,
  oneToOneLine = TRUE, customPar = FALSE, col = "black", lwd = 1,
  cex = 0.8, cex.axis = 1.1, cex.main = 1.1, tinyPlot = FALSE,
  usgsStyle = FALSE, logScale = FALSE, randomCensored = FALSE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotFluxPred(eList)
plotFluxPred(eList, fluxUnit = 'poundsDay')
plotFluxPred(eList, logScale=TRUE)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotFluxPred(eList)

Sample data plot: observed log flux vs log discharge

Description

Concentration and discharge data used to compute flux come from a data frame named Sample which contains the sample data. The metadata come from a data frame named INFO.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Usage

plotFluxQ(eList, qUnit = 2, logScale = TRUE, fluxUnit = 3,
  tinyPlot = FALSE, fluxMax = NA, fluxMin = NA, col = "black",
  lwd = 1, printTitle = TRUE, usgsStyle = FALSE, cex = 0.8,
  cex.axis = 1.1, cex.main = 1.1, customPar = FALSE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotFluxQ(eList, qUnit = 1, fluxUnit = 1)
plotFluxQ(eList, fluxUnit = 'kgDay')
plotFluxQ(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotFluxQ(eList)

Plot of the time series of daily flux estimates and the sample values for the days that were sampled

Description

This plot is useful for visual examination of the ability of the WRTDS, or other model, to fit the data, as seen in a time-series perspective.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotFluxTimeDaily(eList, yearStart = NA, yearEnd = NA, tinyPlot = FALSE,
  fluxUnit = 3, fluxMax = NA, randomCensored = FALSE,
  printTitle = TRUE, plotGenFlux = TRUE, usgsStyle = FALSE, cex = 0.8,
  cex.axis = 1.1, cex.main = 1.1, customPar = FALSE, col = "black",
  lwd = 1, prettyDate = TRUE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotFluxTimeDaily(eList)
plotFluxTimeDaily(eList, 2001,2009)

Makes four graphs of streamflow statistics on a single page

Description

Part of the flowHistory system. The four statistics are 1-day maximum, annual mean, annual 7-day minimum, and the running standard deviation of the log daily discharge values.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Daily dataframe with the daily flow data, and an INFO dataframe with metadata. Each graph shows a loess smooth of the data that are plotted.

Usage

plotFour(eList, yearStart = NA, yearEnd = NA, printTitle = TRUE,
  runoff = FALSE, qUnit = 1, window = 15, cex = 0.8, cex.axis = 1.2,
  cex.main = 1.2, col = "black", lwd = 1, ...)

Arguments

See Also

plotFlowSingle

Examples

eList <- Choptank_eList

#Water year:
plotFour(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList,paStart=6,paLong=3)
plotFour(eList)
 

Makes four graphs of annual streamflow statistics on a single page

Description

Part of the flowHistory system. The four statistics are 1-day maximum, annual mean, annual median, and annual 7-day minimum. Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, a Daily dataframe with the daily flow data, and an INFO dataframe with metadata. Each graph shows a loess smooth of the data that are plotted.

Usage

plotFourStats(eList, yearStart = NA, yearEnd = NA, printTitle = TRUE,
  runoff = FALSE, cex.main = 1.2, qUnit = 1, cex.axis = 1.2,
  cex = 0.8, col = "black", lwd = 1, ...)

Arguments

See Also

plotFlowSingle

Examples

eList <- Choptank_eList

# Water year:
plotFourStats(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList,paStart=6,paLong=3)
plotFourStats(eList)

Plot monthly trend result from runPairs

Description

Plot monthly trend result from runPairs. The change in concentration or flux is calculated from the runPairs function. This plotting function shows an arrow for each month. If the trend from year1 to year2 was increasing, the arrow is red and pointing up. If the trend was decreasing, the arrow is black and pointing down.

Usage

plotMonthTrend(pairResults, yMax = NA, arrowFactor = 0.75, flux = TRUE,
  printTitle = TRUE, concLab = 1, monthLab = 1)

Arguments

Details

The flux values for each month are flow normalized monthly watershed yields expressed as kg/month/km^2. The concentrations are the mean flow normalized concentration, expressed in whatever concentration units the raw data are expressed as (typically mg/L).

Value

Base R plot of monthly trends

Examples

eList <- Choptank_eList
year1 <- 1985
year2 <- 2010



pairOut_1 <- runPairs(eList, year1, year2, windowSide = 0)

plotMonthTrend(pairOut_1)
plotMonthTrend(pairOut_1, flux = FALSE)

eList <- setPA(eList, paStart = 12, paLong = 3)
pairOut_2 <- runPairs(eList, year1, year2, windowSide = 0)

plotMonthTrend(pairOut_2)

eList <- setPA(eList, paStart = 1, paLong = 12)
pairOut_3 <- runPairs(eList, year1, year2, windowSide = 0)

plotMonthTrend(pairOut_3)


 

Plot of the discharge time series

Description

Part of flowHistory component. Allows discharge record to only show those discharges above a given threshold

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotQTimeDaily(eList, yearStart = NA, yearEnd = NA, qLower = NA,
  qUnit = 1, logScale = FALSE, tinyPlot = FALSE, printTitle = TRUE,
  usgsStyle = FALSE, lwd = 3, col = "red", cex.main = 1.2,
  cex.lab = 1.2, customPar = FALSE, prettyDate = TRUE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotQTimeDaily(eList)
plotQTimeDaily(eList, yearStart=1990, yearEnd=2000,qLower=1500)
plotQTimeDaily(eList, prettyDate=FALSE)

Plot of the residuals from WRTDS versus the estimated values (all in log concentration units)

Description

This function produces a plot of the residuals from WRTDS, expressed in natural log concentration units versus the estimated values, also in natural log concentration units. These estimates are the log-space estimates prior to bias-correction. The function provides an alternative for viewing the standardized residuals, where the each residual is divided by its estimated standard error.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Usage

plotResidPred(eList, stdResid = FALSE, tinyPlot = FALSE,
  printTitle = TRUE, col = "black", lwd = 1, cex = 0.8,
  cex.axis = 1.1, cex.main = 1.1, customPar = FALSE,
  randomCensored = FALSE, concLab = 1, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotResidPred(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotResidPred(eList)

Plot of the residuals from WRTDS (in log concentration units) versus the discharge

Description

This function produces a plot of the residuals from WRTDS, expressed in natural log concentration units versus the discharge shown on a log scale. The function also provides an alternative for viewing the standardized residuals, where the each residual is divided by its estimated standard error

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Usage

plotResidQ(eList, qUnit = 2, tinyPlot = FALSE, stdResid = FALSE,
  printTitle = TRUE, col = "black", lwd = 1, cex = 0.8,
  cex.axis = 1.1, cex.main = 1.1, rmSciX = FALSE, customPar = FALSE,
  randomCensored = FALSE, usgsStyle = FALSE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotResidQ(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotResidQ(eList)

Plot of the residuals from WRTDS (in log concentration units) versus time

Description

This function produces a plot of the residuals from WRTDS, expressed in natural log concentration units versus time. It also provides an alternative for viewing the standardized residuals, where the each residual is divided by its estimated standard error.

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Sample dataframe with the sample data, and an INFO dataframe with metadata.

Usage

plotResidTime(eList, stdResid = FALSE, printTitle = TRUE, hLine = TRUE,
  tinyPlot = FALSE, col = "black", lwd = 1, cex = 0.8,
  cex.axis = 1.1, cex.main = 1.1, customPar = FALSE,
  randomCensored = FALSE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList
# Water year:
plotResidTime(eList)
# Graphs consisting of Jun-Aug
eList <- setPA(eList, paStart=6,paLong=3)
plotResidTime(eList)

Graph of the standard deviation of the log of daily discharge versus year

Description

Graph of the standard deviation of the log of daily discharge versus year

Although there are a lot of optional arguments to this function, most are set to a logical default.

Data come from named list, which contains a Daily dataframe with the daily flow data, and an INFO dataframe with metadata.

Usage

plotSDLogQ(eList, yearStart = NA, yearEnd = NA, window = 15,
  sdMax = NA, printTitle = TRUE, tinyPlot = FALSE, printStaName = TRUE,
  printPA = TRUE, cex = 0.8, cex.main = 1.1, cex.axis = 1.1, lwd = 2,
  customPar = FALSE, ...)

Arguments

See Also

selectDays, genericEGRETDotPlot

Examples

eList <- Choptank_eList

# Water year:
plotSDLogQ(eList) 
plotSDLogQ(eList, 1998, 2000) 

plotTimeSlice

Description

Plot of either concentration or flux over time showing both the WRTDS and WRTDSKalman estimates.

Usage

plotTimeSlice(eList, start = NA, end = NA, conc = TRUE, fluxUnit = 3,
  usgsStyle = FALSE)

Arguments

Details

In the plot title, Ratio of means is mean of WRTDSKalman estimates to the WRTDS Classic estimates. Ratio only calculated on the data shown in the figure, not the whole series. In the plot, red dots are measured values, blue dots are plotted at the reporting limit for those values that are censored.

Examples

eList <- Choptank_eList
eList <- WRTDSKalman(eList, niter = 10)

plotTimeSlice(eList, start = 1990, end = 1991, conc = TRUE)

plotTimeSlice(eList, start = 1990, end = 1991, conc = FALSE)

plotTimeSlice(eList, start = NA, end = 1991, conc = FALSE)

plotWRTDSKalman

Description

Two plots to check the flux estimates using the WRTDS_K vs classic WRTDS. The first is annual flux over time, where the two fluxes are shown in different colors. The second is WRTDS vs WRTDSKalman flux estimates. The graphs can be output either on top of each other, or side by side using the sideBySide argument.

Usage

plotWRTDSKalman(eList, sideBySide = FALSE, fluxUnit = 9,
  usgsStyle = FALSE)

Arguments

Examples

eList <- Choptank_eList
eList <- WRTDSKalman(eList, niter = 10)
plotWRTDSKalman(eList)

plotWRTDSKalman(eList, sideBySide = TRUE)

Populate Concentration Columns

Description

Creates ConcLow, ConcHigh, Uncen (0 if censored, 1 if uncensored) columns for Sample data frame for WRTDS analysis.

Usage

populateConcentrations(rawData)

Arguments

Value

concentrationColumns dataframe

Examples

code <- c("","<","")
value <- c(1,2,3)
dataInput <- data.frame(value, code, stringsAsFactors=FALSE)
concentrationDF <- populateConcentrations(dataInput)

Populate Daily data frame

Description

Using raw data that has at least dateTime, value, code, populates the rest of the basic Daily data frame used in EGRET analysis.

Usage

populateDaily(rawData, qConvert, verbose = TRUE)

Arguments

Value

A data frame 'Daily' with the following columns:

Author(s)

Robert M. Hirsch rhirsch@usgs.gov

See Also

readNWISDaily, readUserDaily

Examples

dateTime <- as.character(seq(as.Date("2001/1/1"), 
         as.Date("2001/12/31"), by = "day"))
value <- 1:365
code <- rep("",365)
dataInput <- data.frame(dateTime, value, code, stringsAsFactors=FALSE)
Daily <- populateDaily(dataInput, 2)

Merge concentration to Daily

Description

Used for the WRTDS Kalman set of functions, this function merges the ConcAve into the Daily data frame, renaming it "trueConc", then calculates the "trueFlux", and "stdResid".

Usage

populateDailySamp(eList)

Arguments

Examples

eList <- Choptank_eList
Daily2 <- populateDailySamp(eList)

Populate Date Columns

Description

Creates various date columns for WRTDS study.

Usage

populateDateColumns(rawData)

Arguments

Value

DateFrame dataframe

Examples

dateTime <- c('1984-02-28 13:56',
              '1984-03-01 00:00',
              '1986-03-01 00:00',
              '1986-10-15 00:00')
              
expandedDateDF <- populateDateColumns(dateTime)
expandedDateDF

dateTime <- c('1984-02-28', 
              '1984-03-01',
              '1986-03-01',
              '1986-10-15')
expandedDateDF <- populateDateColumns(dateTime)
expandedDateDF

Populate Parameter Information Columns

Description

Populates INFO data frame with additional user-supplied information concerning the measured parameter.

Usage

populateParameterINFO(parameterCd, INFO, interactive = TRUE)

Arguments

Value

INFO dataframe

Examples

library(dataRetrieval)
INFO <- readNWISsite('01594440')
parameterCd <- "01075"
parameterData <- readNWISpCode(parameterCd)
INFO$param.nm <- parameterData$parameter_nm
INFO$param.units <- parameterData$parameter_units
INFO$paramShortName <- parameterData$srsname
INFO$paramNumber <- parameterData$parameter_cd

INFO <- populateParameterINFO(parameterCd, INFO, interactive = FALSE)

Populate Sample Columns

Description

Creates ConcAve and ConcLow based on Uncen. Removes any samples with NA values in ConcHigh.

Usage

populateSampleColumns(rawData)

Arguments

Value

Sample2 dataframe with columns: Date, ConcLow, ConcHigh, Uncen, ConcAve, Julian, Month, Day, DecYear, MonthSeq, waterYear, SinDY, and CosDY (DY = decimal year)

Examples

dateTime <- c('1985-01-01', '1985-01-02', '1985-01-03')
ConcLow <- c(1,2,0)
ConcHigh <- c(1,2,3)
Uncen <- c(1,1,0)
dataInput <- data.frame(dateTime, ConcLow, ConcHigh, Uncen, stringsAsFactors=FALSE)
Sample <- populateSampleColumns(dataInput)

Populate Site Information Columns

Description

Populates INFO data frame with additional user-supplied information. Also removes fields not related to WRTDS study.

Usage

populateSiteINFO(INFO, siteNumber, interactive = TRUE)

Arguments

Value

INFO dataframe

Examples

library(dataRetrieval)
INFO <- readNWISsite('01594440')
siteNumber <- "01594440"
siteINFO <- populateSiteINFO(INFO, siteNumber, interactive = FALSE)

EGRET helper functions

Description

A small collection of helper functions

Usage

## S3 method for class 'egret'
print(x, ...)

## S3 method for class 'egret'
plot(x, ...)

nDischarge(x)

nObservations(x)

nCensoredVals(x)

Arguments

See Also

multiPlotDataOverview

Examples

Choptank_eList
print(Arkansas_eList)
plot(Choptank_eList)
plot(Choptank_eList, cex.main=0.7)
nDischarge(Arkansas_eList)
nObservations(Arkansas_eList)
nCensoredVals(Arkansas_eList)

Reminder to user of flux unit properties (such as kg/day, tons/year, etc).

Description

Cheat sheet to print out pre-defined flux unit properties from fluxUnit class Flux units included:

Usage

printFluxUnitCheatSheet()

Examples

printFluxUnitCheatSheet()

Print information about group analysis

Description

Prints the information from the runGroups function. This could be used to save the output to a text file.

Usage

printGroups(eList, groupResults)

Arguments

Value

text to console

Examples

eList <- Choptank_eList


groupOut_1 <- runGroups(eList,  windowSide = 0,
                        group1firstYear = 1980, group1lastYear = 1990,
                        group2firstYear = 1995, group2lastYear = 2005)
                       
printGroups(eList, groupOut_1)

Print information about pairs analysis

Description

Prints the information from the runPairs function. This could be used to save the output to a text file.

Usage

printPairs(eList, pairResults)

Arguments

Value

text to console

Examples

eList <- Choptank_eList
year1 <- 1985
year2 <- 2010


pairOut_1 <- runPairs(eList, 
                      year1, year2,
                      windowSide = 0)
                       
printPairs(eList, pairOut_1)

Print annual results for a given streamflow statistic

Description

Part of the flowHistory system. The index of the flow statistics is istat. These statistics are: (1) 1-day minimum, (2) 7-day minimum, (3) 30-day minimum, (4) median (5) mean, (6) 30-day maximum, (7) 7-day maximum, and (8) 1-day maximum.

Usage

printSeries(eList, istat, qUnit = 1, runoff = FALSE)

Arguments

Details

The smoothing algorithm is as defined in makeAnnualSeries. smoothing window is defined by the eList$INFO$window value (default = 20)

Value

data frame with:

Examples

eList <- Choptank_eList
printReturn <- printSeries(eList, 5)

Reminder to user of flow Unit properties such as cubic meters per second or thousands of cubic feet per second.

Description

Cheat sheet to print out pre-defined qUnit properties from qUnit class. Flow units included:

Usage

printqUnitCheatSheet()

Examples

printqUnitCheatSheet()

Processing of Water Quality Data

Description

Processes water quality data. This function looks at detection limit and detection conditions to determine if a value is left censored or not. Censored values are given the qualifier "<". The dataframe is also converted from a long to wide format.

Usage

processQWData(data)

Arguments

Value

data dataframe with first column dateTime, and at least one qualifier and value columns (subsequent qualifier/value columns could follow depending on the number of parameter codes)

See Also

readWQPqw

Examples

#rawWQP <- dataRetrieval::readWQPqw('21FLEECO_WQX-IMPRGR80','Phosphorus', '', '')
#Sample2 <- processQWData(rawWQP)

qUnit class

Description

Some details about the qUnit class

Details

qshortName

A character specifying the short name.

qUnitFactor

A numeric representing the conversion factor

qUnitName

A character specifying the full name.

qUnitExpress

An expression specifying the full name.

unitUSGS

A character specifying flux with full text.

qUnitTiny

An expression specifying the abbreviated name.

shortCode

A number for quick lookup

prefix

A character specifying the general type of measurement.

randomSubset

Description

Calculates a random subset of the data based on repeated values from a specified column.

Usage

randomSubset(df, colName, seed = NA)

Arguments

Examples

df <- data.frame(Julian = c(1,2,2,3,4,4,4,6),
                 y = 1:8)
df
df_random <- randomSubset(df, "Julian")
df_random

Basic Data Import for Water Flow Data

Description

Imports data from user-supplied data file. Specifically used to import water flow data for use in the EGRET package. For EGRET usage, the first column is expected to be dates. If the data is daily data, then next column is expected to be the measured values. If the data is sampled data, the next column is remark codes, and the third column is values. If the date format is not automatically detected, the format can be specified using the "format" argument.

Usage

readDataFromFile(filePath, fileName, hasHeader = TRUE, separator = ",",
  format = "%m/%d/%Y")

Arguments

Value

retval dataframe with dateTime, value, and code columns

Examples

filePath <- system.file("extdata", package="EGRET")
fileName <- 'ChoptankRiverFlow.txt'
ChopData <- readDataFromFile(filePath,fileName, separator="\t")

Import NWIS Daily Data for EGRET analysis

Description

Imports daily data from NWIS web service. This function gets the data from here: https://waterservices.usgs.gov/

Usage

readNWISDaily(siteNumber, parameterCd = "00060", startDate = "",
  endDate = "", verbose = TRUE, convert = TRUE)

Arguments

Value

A data frame 'Daily' with the following columns:

See Also

readNWISdv, populateDaily

Examples

Daily <- readNWISDaily('01594440','00060',
                       '1985-01-01', '1985-03-31')
DailySuspSediment <- readNWISDaily('01594440','80154',
                                   '1985-01-01', '1985-03-31',
                                   convert = FALSE)

Import USGS Sample Data for EGRET analysis

Description

Imports data from USGS web service. For raw data, use read_USGS_samples from the dataRetrieval package. This function will retrieve the raw data, and compress it (summing constituents) if more than 1 parameter code is supplied. See section 3.2.4 of the vignette for more details.

Usage

readNWISSample(siteNumber, parameterCd, startDate = "", endDate = "",
  verbose = TRUE)

Arguments

Value

A data frame 'Sample' with the following columns:

See Also

compressData, populateSampleColumns, readWQPqw

Examples

# These examples require an internet connection to run

Sample_01075 <- readNWISSample('01594440','01075', '1985-01-01', '1985-03-31')

Import user daily data for EGRET analysis

Description

Imports data from a user-supplied file, and converts it to a Daily data frame, appropriate for WRTDS calculations. The file can use most any separators as a delimiter. The default is comma separated.

The first column is expected to be dates, the second column is expected to be discharge values. If the date format is not automatically detected, the format can be specified using the "format" argument.

Usage

readUserDaily(filePath, fileName, hasHeader = TRUE, separator = ",",
  qUnit = 1, format = "%m/%d/%Y", verbose = TRUE)

Arguments

Value

A data frame 'Daily' with the following columns:

Examples

filePath <- system.file("extdata", package="EGRET")
fileName <- "ChoptankRiverFlow.txt"
Daily <- readUserDaily(filePath,fileName,separator="\t")

Import user-supplied sample data for EGRET analysis

Description

Imports data from a user-supplied file, and converts it to a Sample data frame (including summing multiple constituents), appropriate for EGRET analysis. First column is date, second is remark code, and third is value. If multiple constituents are to be combined with interval censoring, additional pairs of columns can be inserted, each pair starting with remark code (specifically looking for <), followed by the values. If the date format is not automatically detected, the format can be specified using the "format" argument.

Usage

readUserSample(filePath, fileName, hasHeader = TRUE, separator = ",",
  verbose = TRUE, format = "%m/%d/%Y")

Arguments

Value

A data frame 'Sample' with the following columns:

See Also

compressData, populateSampleColumns

Examples

filePath <- system.file("extdata", package="EGRET")
fileName <- 'ChoptankRiverNitrate.csv'
Sample <- readUserSample(filePath,fileName, separator=";",verbose=FALSE)

Import Sample Data from the Water Quality Portal for WRTDS

Description

Imports data from the Water Quality Portal, so it could be STORET, USGS, or USDA data. This function gets the data from: https://www.waterqualitydata.us For raw data, use readWQPdata. This function will retrieve the raw data, compress it (summing constituents), then converts it to the Sample dataframe structure. See chapter 7 of the EGRET user guide for more details.

Usage

readWQPSample(siteNumber, characteristicName, startDate = "", endDate = "",
  verbose = TRUE, legacy = TRUE)

Arguments

Value

A data frame 'Sample' with the following columns:

See Also

readWQPdata, dataRetrieval::whatWQPsites, readWQPqw, compressData, populateSampleColumns

Examples

# These examples require an internet connection to run

Sample_All <- readWQPSample('WIDNR_WQX-10032762','Specific conductance', '', '')

Remove duplicates values from Sample data frame.

Description

Removes observations from the data frame Sample when the observation has the identical date and value as another observation

Usage

removeDuplicates(Sample)

Arguments

Value

A data frame 'Sample' with the following columns:

Examples

DecYear <- c('1985.01', '1985.01', '1985.02', '1985.02', '1985.03')
ConcHigh <- c(1,2,3,3,5)
dataInput <- data.frame(DecYear, ConcHigh, stringsAsFactors=FALSE)
Sample <- removeDuplicates(dataInput)

Runs a comparison of any group of years in the record.

Description

runGroups provides comparisons of results, in terms of flow-normalized concentration and flow-normalized flux for any groups of years of years in the water quality record. Comparison could involve the use of the "wall" and/or use of "generalized flow-normalization". These two concepts are described in detail in the vignette: vignette("Enhancements", package = "EGRET").

Usage

runGroups(eList, windowSide, group1firstYear, group1lastYear, group2firstYear,
  group2lastYear, surfaceStart = NA, surfaceEnd = NA, flowBreak = FALSE,
  Q1EndDate = NA, QStartDate = NA, QEndDate = NA, wall = FALSE,
  oldSurface = FALSE, fractMin = 0.75, sample1EndDate = NA,
  sampleStartDate = NA, sampleEndDate = NA, paStart = NA, paLong = NA,
  minNumObs = 100, minNumUncen = 50, windowY = 7, windowQ = 2,
  windowS = 0.5, edgeAdjust = TRUE, verbose = TRUE, saveOutput = FALSE,
  fileName = "temp.txt")

Arguments

Details

When using generalized flow-normalization, it is best to have the Daily data frame extend well beyond the years that are in the Sample data frame. Ideally, the Daily data frame would start windowSide years before the start of the Sample data set, if the data exist to provide for that. Generally that isn't possible for the end of the record because the Sample data may end very close to the present. To the extent that is possible therefore, it is better to include more discharge data after the end of the Sample record. Also note that in the case run in the examples don't do that, because the data set needs to be appropriate for stationary flow normalization as well (and package size considerations make it difficult to include specialized examples).

Value

Dataframe with 7 columns and 2 rows. The first row is about trends in concentration (mg/L), the second column is about trends in flux (million kg/year). The data frame has a number of attributes.

Examples

eList <- Choptank_eList


#Option 1:  Use all years for group flow normalization.
groupOut_1 <- runGroups(eList,  windowSide = 0,
                       group1firstYear = 1980, group1lastYear = 1990,
                       group2firstYear = 1995, group2lastYear = 2005)

# Option 2: Use sliding window.
#                In each case it is a 23 year window (23 = 1 + 2 * 11)

groupOut_2 <- runGroups(eList,  windowSide = 11,
                       group1firstYear = 1980, group1lastYear = 1990,
                       group2firstYear = 1995, group2lastYear = 2005)

# Option 3: Flow normalization is based on splitting the flow record at 1990-09-30
#                But in years before the break it uses all flow data from before the break,
#                and years after the break uses all flow data after the break

groupOut_3 <- runGroups(eList,  windowSide = 0,
                       group1firstYear = 1980, group1lastYear = 1990,
                       group2firstYear = 1995, group2lastYear = 2005,
                       flowBreak = TRUE, 
                       Q1EndDate = "1990-09-30")

# Option 4: Flow normalization is based on splitting the flow record at 1990-09-30
#                but before the break uses a 23 year window of years before the break
#                after the break uses a 23 year window of years after the break
groupOut_4 <- runGroups(eList,  windowSide = 11,
                       group1firstYear = 1980, group1lastYear = 1990,
                       group2firstYear = 1995, group2lastYear = 2005,
                       flowBreak = TRUE, 
                       Q1EndDate = "1990-09-30")

Runs a comparison of any two years in the record.

Description

runPairs provides comparisons of results, in terms of flow-normalized concentration and flow-normalized flux for any pair of years in the water quality record. Comparison could involve the use of the "wall" and/or use of "generalized flow normalization". These two concepts are described in detail in the vignette: vignette("Enhancements", package = "EGRET").

Usage

runPairs(eList, year1, year2, windowSide, flowBreak = FALSE,
  Q1EndDate = NA, QStartDate = NA, QEndDate = NA, wall = FALSE,
  oldSurface = FALSE, sample1EndDate = NA, sampleStartDate = NA,
  sampleEndDate = NA, paStart = NA, paLong = NA, minNumObs = 100,
  minNumUncen = 50, fractMin = 0.75, windowY = 7, windowQ = 2,
  windowS = 0.5, edgeAdjust = TRUE, saveOutput = FALSE,
  fileName = "temp.txt", verbose = TRUE)

Arguments

Details

When using generalized flow-normalization, it is best to have the Daily data frame extend well beyond the years that are in the Sample data frame. Ideally, the Daily data frame would start windowSide years before the start of the Sample data set, if the data exist to provide for that. Generally that isn't possible for the end of the record because the Sample data may end very close to the present. To the extent that is possible therefore, it is better to include more discharge data after the end of the Sample record. Also note that in the case run in the examples don't do that, because the data set needs to be appropriate for stationary flow normalization as well (and package size considerations make it difficult to include specialized examples).

Value

Data frame with 7 columns and 2 rows. The first row is about trends in concentration (mg/L), the second column is about trends in flux (million kg/year). The data frame has a number of attributes.

Additionally, there is an attribute on the data frame "Other", containing a list that includes minNumObs=minNumObs, minNumUncen, windowY, windowQ, siteName, windowS, wall, edgeAdjust, QStartDate, QEndDate, PercentChangeConc, and PercentChangeFlux.

PercentChangeConc, and PercentChangeFlux are vectors where: Total Percent Change is the Total Change divided by x11 CQTC Percent is the CQTC divided by x11 QTC Percent is the QTC divided by x11

Another attribute is "byMonth". This is a data frame of 4 columns and 14 rows. The columns represent the concentrations and fluxes for the starting and ending year. The flux values for each month are flow normalized monthly watershed yields expressed as kg/month/km^2. The concentrations are the mean flow normalized concentration, expressed in whatever concentration units the raw data are expressed as (typically mg/L). This data frame is used as the input to the plotMonthTrend function.

Examples

eList <- Choptank_eList
year1 <- 1985
year2 <- 2010


# Automatic calculations based on windowSide = 11
# four possible ways to do generalized flow normalization:

#Option 1: Use all years for flow normalization.

pairOut_1 <- runPairs(eList, year1, year2, windowSide = 0)

# Option 2:  Use different windows for flow normalization for year1 versus year2
#            In each case it is a 23 year window (23 = 1 + 2*11)

pairOut_2 <- runPairs(eList, year1, year2, windowSide = 11)

# Option 3: Flow normalization is based on splitting the flow record at 1990-09-30
#          But year1 uses all flow data from before the break, 
#          year2 uses all flow data after the break

pairOut_3 <- runPairs(eList, year1, year2, 
                      windowSide = 0, flowBreak = TRUE,
                      Q1EndDate = "1990-09-30")

# Option 4: Flow normalization is based on splitting the flow record at 1990-09-30
#           but year1 uses a 23 year window before the break
#           year2 uses a 23 year window after the break

pairOut_4 <- runPairs(eList, year1, year2, 
                      windowSide = 11, flowBreak = TRUE,
                      Q1EndDate = "1990-09-30")
                      
monthly_trends <- attr(pairOut_4, "byMonth")  
plotMonthTrend(pairOut_4)

eList <- setPA(eList, paLong = 3, paStart = 12)
pairOut_5 <- runPairs(eList, year1, year2,
                      windowSide = 11)
monthly_trends <- attr(pairOut_5, "byMonth") 
plotMonthTrend(pairOut_5)                      

Annual series of flow-normalized concentration and flow-normalized flux

Description

runSeries provides annual series of flow-normalized concentration and flow-normalized flux for the water quality record. Computations could involve the use of the "wall" and/or use of "generalized flow normalization". These two concepts are described in detail in the vignette: vignette("Enhancements", package = "EGRET").

Usage

runSeries(eList, windowSide, surfaceStart = NA, surfaceEnd = NA,
  flowBreak = FALSE, Q1EndDate = NA, QStartDate = NA, QEndDate = NA,
  wall = FALSE, oldSurface = FALSE, sample1EndDate = NA,
  sampleStartDate = NA, sampleEndDate = NA, paStart = NA, paLong = NA,
  fractMin = 0.75, minNumObs = 100, minNumUncen = 50, windowY = 7,
  windowQ = 2, windowS = 0.5, edgeAdjust = TRUE, verbose = TRUE)

Arguments

Details

When using generalized flow-normalization, it is best to have the Daily data frame extend well beyond the years that are in the Sample data frame. Ideally, the Daily data frame would start windowSide years before the start of the Sample data set, if the data exist to provide for that. Generally that isn't possible for the end of the record because the Sample data may end very close to the present. To the extent that is possible therefore, it is better to include more discharge data after the end of the Sample record. Also note that in the case run in the examples don't do that, because the data set needs to be appropriate for stationary flow normalization as well (and package size considerations make it difficult to include specialized examples).

Value

eList named list with INFO, Daily, and Sample dataframes, along with the surfaces matrix.

Examples

eList <- Choptank_eList


# Automatic calculations based on windowSide = 11
# four possible ways to do generalized flow normalization

#Option 1:  Use all years for flow normalization.
seriesOut_1 <- runSeries(eList,  windowSide = 0)
plotConcHist(seriesOut_1)
plotFluxHist(seriesOut_1)

# Option 2: Use sliding window throughout the whole flow normalization process.
#                In each case it is a 15 year window (23 = 1 + 2*11)
seriesOut_2 <- runSeries(eList, windowSide = 11)

plotConcHist(seriesOut_2)
plotFluxHist(seriesOut_2)

# Option 3: Flow normalization is based on splitting the flow record at 1990-09-30
#                But in years before the break it uses all flow data from before the break, 
#                and years after the break uses all flow data after the break
seriesOut_3 <- runSeries(eList,
                       windowSide = 0, 
                       flowBreak = TRUE,
                       Q1EndDate = "1990-09-30")
                       
plotConcHist(seriesOut_3)
plotFluxHist(seriesOut_3)

# Option 4: Flow normalization is based on splitting the flow record at 1990-09-30
#                but before the break uses a 23 year window of years before the break
#                after the break uses a 23 year window of years after the break
seriesOut_4 <- runSeries(eList, 
                      windowSide = 11, flowBreak = TRUE,
                      Q1EndDate = "1990-09-30")
                      
plotConcHist(seriesOut_4)
plotFluxHist(seriesOut_4)

Run the weighted survival regression for a set of estimation points (defined by DecYear and Log(Q))

Description

This function runs the survival regression which is the concentration estimation method of WRTDS. It uses sample data from the data frame Sample. It does the estimation for a set of data points defined by two vectors: estPtYear and estPtLQ. It returns an array of results for the estimation points. The array returned contains yHat, SE and ConcHat (in that order). yHat is the expected value of log(concentration), SE is the standard error of log(concentration) and ConcHat is the expected value of concentration.

Usage

runSurvReg(estPtYear, estPtLQ, DecLow, DecHigh, Sample, windowY = 7,
  windowQ = 2, windowS = 0.5, minNumObs = 100, minNumUncen = 50,
  verbose = TRUE, interactive = NULL, edgeAdjust = TRUE,
  run.parallel = FALSE)

run_WRTDS(estY, estLQ, localSample, DecLow, DecHigh, minNumObs, minNumUncen,
  windowY, windowQ, windowS, edgeAdjust)

Arguments

Value

resultSurvReg numeric array containing the yHat, SE, and ConcHat values array dimensions are (numEstPts,3)

Examples

eList <- Choptank_eList
estPtYear<-c(2001.0,2005.0,2009.0)
estPtLQ<-c(1,1,1)
Sample <- getSample(eList)
DecLow <- Sample$DecYear[1]
DecHigh <- Sample$DecYear[nrow(Sample)]
resultSurvReg <- runSurvReg(estPtYear,estPtLQ,
                            DecLow,DecHigh,Sample,
                            run.parallel = FALSE)

A utility program for saving the contents of the workspace This function saves the workspace. Future versions of EGRET will not include this function, use saveRDS to save individual eList objects. It assigns the file a name using the abbreviations for station and constituent.

Description

A utility program for saving the contents of the workspace

This function saves the workspace. Future versions of EGRET will not include this function, use saveRDS to save individual eList objects. It assigns the file a name using the abbreviations for station and constituent.

Usage

saveResults(savePath, eList)

Arguments

Creates a subset Daily data frame that only contains daily estimates for the specified period of analysis

Description

This function uses the user-defined 'period of analysis', and subsets the Daily data frame, it doesn't have any effect on the Sample data frame. If you want to examine your data set as a time series of water years, then the period of analysis is October through September. If you want to examine the data set as calendar years then the period of analysis is January through December. You might want to examine the winter season, which you could define as December through February, then those 3 months become the period of analysis. The only constraints on the definition of a period of analysis are these: it must be defined in terms of whole months; it must be a set of contiguous months (like March-April-May), and have a length that is no less than 1 month and no more than 12 months. Define the PA by using two arguments: paLong and paStart. paLong is the length of the period of analysis, and paStart is the starting month.

Usage

selectDays(df, paLong, paStart)

Arguments

Value

localDaily a data frame containing the daily data but only for the period of analysis (not all months)

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)
DailySubset <- selectDays(Daily, 4, 11)

Sets up the period of analysis (the portion of the year being evaluated).

Description

Period of analysis is defined by the starting month (paStart) and length in months (paLong). paStart and paLong are constrained to be integers from 1 to 12. For example, a water year would be paStart = 10 and paLong = 12. For example, the winter season, defined by Dec,Jan,Feb would be paStart = 12 and paLong = 3.

Usage

setPA(eList, paStart = 10, paLong = 12, window = 20)

Arguments

Value

eList named list with at least the INFO dataframe. Any of these values can be NA, but not all EGRET functions will work with missing parts of the named list eList.

Examples

eList <- Choptank_eList
eList <- setPA(eList, paStart = 12, paLong = 3)

Create a character variable that describes the period of analysis, when period of analysis has already been set in AnnualResults

Description

The period of analysis can be of any length from 1 month to 12 months. The period of analysis can have any starting month from 1 (January) through 12 (December). This function produces a character character that describes this period of analysis. For example "water year", "calendar year", "year starting with April", or "Season consisting of April, May, June". There is an alternative version of this function for the case where AnnualResults does not exist. This might arise in a call from plotConcTime or plotLogConcTime. That function is called setSeasonLabelByUser.

Usage

setSeasonLabel(localAnnualResults, monthLab = 1)

Arguments

Value

periodName character which describes the period of analysis

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)
AnnualResults <- setupYears(Daily)
setSeasonLabel(AnnualResults)

AnnualResultsWinter <- setupYears(Daily, 
                                  paLong = 3,
                                  paStart = 12)
setSeasonLabel(AnnualResultsWinter)

Creates a character variable that describes the period of analysis, when the period of analysis is being set by the user and not from AnnualResults

Description

The period of analysis can be of any length from 1 month to 12 months. The period of analysis can have any starting month from 1 (January) through 12 (December). This function produces a character that describes this period of analysis. For example "water year", "calendar year", "year starting with April", or "Season consisting of April, May, June". There is an alternative version of this function for the case where AnnualResults exists. And we want to use the period of analysis defined there. That function is called setSeasonLabel.

Usage

setSeasonLabelByUser(paStartInput = 10, paLongInput = 12, monthLab = 1)

Arguments

Value

periodName character which describes the period of analysis

Examples

setSeasonLabelByUser(paStartInput=1,paLongInput=12)
setSeasonLabelByUser(paStartInput=4,paLongInput=3)

setUpEstimation

Description

Set up the INFO data frame for a modelEstimation

Usage

setUpEstimation(eList, windowY = 7, windowQ = 2, windowS = 0.5,
  minNumObs = 100, minNumUncen = 50, edgeAdjust = TRUE, verbose = TRUE,
  interactive = NULL)

Arguments

Value

eList named list with Daily, Sample, and INFO dataframes.

Examples

eList <- Choptank_eList
eList <- setUpEstimation(eList)

Creates the AnnualResults data frame from the Daily data frame

Description

This function aggregates the results stored on a daily basis in the Daily data frame and stores the average values of these in the new data frame called AnnualResults. Note that the flux values are rates (kg/day) and not a mass (kg). The "annual values" can be a full 12 months, or they can be shorter. See manual to understand paLong and paStart arguments. The simplest case, a Water Year (October through September), would have paLong=12, and paStart=10. A calendar year would be paLong=12 and paStart=1. A winter season of Dec, Jan, Feb would be paLong=3 and paStart=12

Usage

setupYears(localDaily, paLong = 12, paStart = 10)

Arguments

Value

A data frame 'AnnualResults' of numeric values with the following columns

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)
AnnualResults <- setupYears(Daily, 4, 10)

startEnd

Description

Returns two date variables representing the starting date and ending date for a combination of paStart, paLong, and year

Usage

startEnd(paStart, paLong, year)

Arguments

Value

Date list

Examples

paStart <- 10
paLong <- 12
year <- 1999
startEnd(paStart, paLong, year)

stitch surfaces

Description

This function creates a continuous surfaces object that starts just before surfaceStart and ends just after surfaceEnd. It is made up from two surfaces objects created when there is a wall specified for the analysis. The first surfaces object is based on data prior to the wall and the second surfaces object is based on data after the wall. The wall is located just after sample1EndDate. The Daily data frame is used only to set the minimum and maximum discharges used to construct the indices for discharges in the surfaces.

Usage

stitch(eList, sample1StartDate, sample1EndDate, sample2StartDate,
  sample2EndDate, surfaceStart = NA, surfaceEnd = NA, minNumObs = 100,
  minNumUncen = 50, fractMin = 0.75, windowY = 7, windowQ = 2,
  windowS = 0.5, edgeAdjust = TRUE, verbose = FALSE,
  run.parallel = FALSE)

Arguments

Examples

eList <- Choptank_eList

surfaceStart <- "1986-10-01"
surfaceEnd <- "2010-09-30"

# Surface skips a few years:
sample1StartDate <- "1986-10-01"
sample1EndDate <- "1992-09-30"
sample2StartDate <- "1996-10-01"
sample2EndDate <- "2011-09-30"


surface_skip <- stitch(eList, 
                         sample1StartDate, sample1EndDate, 
                         sample2StartDate, sample2EndDate,
                         surfaceStart, surfaceEnd)

# Surface overlaps a few years:
sample1StartDate <- "1986-10-01"
sample1EndDate <- "1996-09-30"
sample2StartDate <- "1992-10-01"
sample2EndDate <- "2011-09-30"

surface_overlap <- stitch(eList, 
                         sample1StartDate, sample1EndDate, 
                         sample2StartDate, sample2EndDate)

Compute the 6 parameters needed to lay out the grid for the surfaces computed in estSurfaces

Description

The code here is a repetition of the first part of the code for estSurfaces

Usage

surfaceIndex(Daily)

Arguments

Value

surfaceIndexParameters a numeric vector of length 6, defining the grid for the surfaces

Examples

eList <- Choptank_eList
Daily <- getDaily(eList)
surfaceIndex(Daily)

Surface date limits

Description

Sets the Date limits to the surfaces being estimated from the Sample data set. The start is less than a year prior to the first date (typically the date of the first sample) and the end is less than a year after the last date (typically the date of the last sample). The start is constrained to be on the first day of the period of analysis and the end is constrained to be on the last day of the the period of analysis

Usage

surfaceStartEnd(paStart, paLong, Date1, Date2)

Arguments

Examples

eList <- Choptank_eList
Date1 <- eList$Sample$Date[1]
Date2 <- range(eList$Sample$Date)[2]
surfaceStartEnd(10, 12, Date1, Date2)

Create a table of the changes in flow-normalized values between various points in time in the record

Description

These tables describe trends in flow-normalized concentration and in flow-normalized flux. They are described as changes in real units or in percent and as slopes in real units per year or in percent per year. They are computed over pairs of time points. These time points can be user-defined or they can be set by the program to be the final year of the record and a set of years that are multiples of 5 years prior to that. tableChangeSingle is a version of the same code that will produce output only for flow-normalized concentration or flow-normalized flux, but not both

Usage

tableChange(eList, fluxUnit = 9, yearPoints = NA)

tableChangeSingle(eList, fluxUnit = 9, yearPoints = NA, flux = FALSE)

Arguments

Value

dataframe with Year1, Year2, change[mg/L], slope[mg/L], change[percent], slope[percent] columns. The data in each row is the change or slope calculated from Year1 to Year2

Examples

eList <- Choptank_eList
# Water Year:

tableChange(eList, fluxUnit = 8, yearPoints = c(1980, 1995, 2011))
tableChange(eList, fluxUnit = 5) 
# Winter:
eList <- setPA(eList, paStart = 12, paLong = 3)
tableChange(eList, fluxUnit = 8, yearPoints = c(1980, 1995, 2011))

# Water Year:
eList <- setPA(eList, paStart = 10, paLong = 12)
#This returns concentration ASCII table in the console:
tableChangeSingle(eList, fluxUnit = 8, yearPoints = c(1980, 1995, 2011), flux = FALSE)
#Returns a data frame:
change <- tableChangeSingle(eList, fluxUnit = 8, yearPoints=c(1980, 1995, 2011), flux = FALSE) 
#This returns flux values as a data frame:
df <- tableChangeSingle(eList, fluxUnit = 8, yearPoints=c(1980, 1995, 2011), flux = TRUE)  
# Winter Concentration only:
eList <- setPA(eList, paStart = 12, paLong = 3)
df.winter <- tableChangeSingle(eList, fluxUnit = 8, yearPoints=c(1980, 1995, 2011), flux = FALSE)

Prints table of change metrics for a given streamflow statistic

Description

Part of the flowHistory system. Provides a measure of change (in real units and as percent per year) based on the smoothed values for various streamflow statistics. Smoothing algorithm is the same as is used in plotFlowSingle.

Usage

tableFlowChange(eList, istat, qUnit = 1, runoff = FALSE, yearPoints = NA)

Arguments

Details

The index of the flow statistics is istat. These statistics are: (1) 1-day minimum, (2) 7-day minimum, (3) 30-day minimum, (4) median (5) mean, (6) 30-day maximum, (7) 7-day maximum, and (8) 1-day maximum.

Can also run the statistics on any Period of Analysis (individual months or sequence of months) using setPA.

A dataframe is returned, as well as a printout in the R console.

Examples

eList <- Choptank_eList
tableFlowChange(eList, istat = 5, yearPoints = c(1981, 1995, 2010))
eList <- setPA(eList, paStart = 4, paLong = 12)
tableFlowChange(eList, istat = 2, qUnit = 2, yearPoints = c(1981, 1995, 2010))
eList <- setPA(eList, paStart = 9, paLong = 1)
df <- tableFlowChange(eList, istat = 5, qUnit = 2, yearPoints = c(1981, 1995, 2010))
df 

Table of annual results for discharge, concentration and flux

Description

Produce an ASCII table showing: year, mean discharge, mean concentration, flow-normalized concentration, mean flux, and flow-normalized flux. Note that the flux and flow-normalized flux are rates and not a mass. As such a value for some period shorter than a full year could be larger than the value for a full year.

Usage

tableResults(eList, qUnit = 2, fluxUnit = 9, localDaily = NA)

Arguments

Details

Can also procude a table for any Period of Analysis (individual months or sequence of months) using setPA.

Value

results dataframe, if returnDataFrame=TRUE

dataframe with year, discharge, concentration, flow-normalized concentration, flux, and flow-normalized concentration columns. If the eList was run through WRTDSKalman, an additional column generalized flux is included.

Examples

eList <- Choptank_eList
# Water Year:

tableResults(eList, fluxUnit = 8)
df <- tableResults(eList, fluxUnit = 1)
df
# Spring:
eList <- setPA(eList, paStart = 3, paLong = 3)
tableResults(eList, fluxUnit = 1, qUnit = "cfs")

Tricube weight function

Description

Computes the tricube weight function on a vector of distance values (d), based on a half-window width of h, and returns a vector of weights that range from zero to 1.

Usage

triCube(d, h)

Arguments

Details

See Cleveland, W. S. (1979). Robust locally weighted regression and smoothing scatterplots, JASA, 74, 829-836

Value

w numeric vector of weights, all 0<=w<=1

Examples

 h<-10
 d<-c(-11,-10,-5,-1,-0.01,0,5,9.9,10,20)
 triCube(d,h)

Sets up tick marks for an axis for a graph with an arithmetic scale which starts at zero

Description

Axis tick marks that run from zero to some specified maximum, creates about 4 to 8 ticks marks.

Usage

yPretty(yMax)

Arguments

Value

yTicks A numeric vector representing the values for each of the tick marks

Examples

yTicks <- yPretty(7.8)
yTicks <- yPretty(125)